CN1324617C - Formation of thin film resistors - Google Patents

Abstract

The present invention relates to thin film resistors which may be embedded in multilayer printed wiring boards, to the structures used to form these thin film resistors and to the methods of forming these structures, including the use of combustion chemical vapor deposition. The invention also relates to chemical precursor solutions with which resistive materials can be deposited on substrates by combustion chemical vapor deposition techniques.

Description

Formation of thin film resistors

This application is a divisional application of the application "Formation of Thin Film Resistance" with application number 99106358.9 and application date of April 29, 1999.

technical field

The present invention is directed to the formation of sheet resistors, preferably sheet resistors for printed circuits, such thin layers that can be embedded in printed wiring boards. In particular, the present invention is directed to forming sheet resistors from sheet resistive materials that can be deposited by combustion chemical vapor deposition.

Background technique

Combustion chemical vapor deposition ("CCVD") is a recently invented CVD technique that allows the deposition of thin films in an open atmosphere. The CCVD process has several advantages over other thin film technologies, including conventional CVD techniques. The main advantage of CCVD is that it can deposit films in an open atmosphere without any expensive furnaces, vacuum devices or reaction chambers. As a result, the initial system investment can be reduced by up to 90% compared to vacuum-based systems. Instead of the specialized environment required by other techniques, the combustion flame provides the environment needed to deposit the essential components from a solution, vapor or gas source. The precursor is usually dissolved in a solvent, which is also used as a combustible fuel. Deposition can be performed at atmospheric pressure and temperature, in an exhaust hood, outdoors, or in a chamber with a controlled ambient gas or pressure.

Because CCVD typically uses solutions, a significant advantage of this technique is its rapid and simple change of dopants and stoichiometry, which facilitates the deposition of composite films. CCVD techniques typically use inexpensive, soluble precursors. Furthermore, the vapor pressure of the precursors does not play an important role in CCVD because the dissolution process provides the energy for the generation of the necessary ionic species. By adjusting solution concentration and composition, a wide range of stoichiometry can be deposited quickly and easily. Furthermore, the CCVD process makes it possible to tailor the chemical composition and physical structure of the deposited film to the requirements of a specific application.

Unlike conventional CVD, CCVD processes are not limited to expensive, fixed, low-pressure reactors. Thus, the deposition flame or flame beam can be moved along the substrate to readily coat large and/or composite surface areas. Because the CCVD process is not limited to a dedicated environment, the user can continuously add material to the coating area without interruption, thus enabling batch processing. Furthermore, the user can limit deposition to specific areas of the substrate by simply controlling the residence time of the flame on those areas. Finally, CCVD techniques generally use halogen-free chemical precursors, which greatly reduces the negative impact on the environment.

A variety of materials have been deposited by CCVD techniques using the combustion of premixed precursor solutions as the sole heat source. This cheap and flexible film deposition technique can make thin-film technology widely applicable. The CCVD process has roughly the same flexibility as thermal spraying, while producing conformal films of the same quality as those produced by conventional CVD. Using CCVD processes, the desired phases can be deposited within days at a relatively low cost.

A preferred embodiment of the CCVD process is described in detail in US Application No. 08/691,853, filed August 2, 1996, the contents of which are incorporated herein by reference. According to this application, CCVD produces gas-phase formed films, powders and nanophase coatings from near-supercritical liquids and supercritical fluids. It is preferred to form a liquid or liquid-like solution fluid containing one or more chemical precursors. The solution fluid is conditioned to near or above the critical pressure, then heated to near the supercritical temperature, and released through a restrictor or nozzle to produce a gas entraining a very fine atomized or vaporized solution fluid. The solution fluid vapor is combusted to form a flame or passed into a flame or electric flame plasma where the precursors react to the desired phases or on the surface of the substrate. Due to the high temperature of the plasma, many precursors react before the substrate surface. The coating is deposited by placing the substrate in or near a flame or torch plasma. Furthermore, the formed material can be collected as a nanophase powder.

Very fine atomization, mist, vaporization or vaporization is obtained using solution fluids at or above the critical pressure, near the critical temperature. The dissolved chemical precursor need not have a high vapor pressure, but a precursor with a high vapor pressure is preferred over a precursor with a lower vapor pressure. By heating the solution fluid just before or at the end of the nozzle or restrictor (atomization device), the time available for chemical reaction or dissolution of the precursor prior to atomization is minimized. The method can be used to deposit coatings from a variety of organometallic compounds and inorganic precursors. The solvent of the fluid solution may be selected from any liquid or supercritical fluid in which the precursors can form a solution. The liquid or fluid solvent itself may consist of a mixture of different compounds.

Lowering the supercritical temperature of the reagent-containing fluid results in superior coatings. Many of these fluids are not stable liquids at STP (Standard Temperature Pressure) and must be mixed in a pressure cylinder or at cryogenic temperatures. To facilitate the formation of a liquid or fluid solution that can only exist at pressures above ambient pressure, one or more chemical precursors may optionally be first dissolved in a first solvent that is stable at ambient pressure. This solution is placed in a container capable of withstanding pressure, and a second (primary) liquid or fluid (in which the first solution is miscible) is added. The supercritical temperature of the primary liquid or fluid is lower, thereby reducing the maximum temperature required to produce the desired level of spray. By forming a high concentration first solution, many of the resulting lower concentration solutions are composed of second solution compounds and possibly additional solution compounds. In general, the higher the proportion of a given compound in a given solution, the more the solution behaves like that compound. These added liquids and fluids are selected to facilitate very fine atomization, vaporization or vaporization of the chemical precursor-containing solution. Choosing a final solution mixture with a low supercritical temperature also minimizes the reaction of chemical precursors inside the atomization device and reduces or eliminates the need to heat the solution in the release area. In some instances, the solution may be cooled prior to the release zone to maintain solubility and fluid stability. Those skilled in the art of supercritical fluid solutions can identify many possible solution mixtures without undue experimentation. Optionally, a pressure vessel with glass windows or a pressure vessel with fiber optics and monitors enables visual determination of miscibility and solute-solvent compatibility. Conversely, if the in-line filter is clogged or if sediment is found to remain in the main vessel, an incompatibility may have occurred under those conditions.

Another advantage is the rapid expansion of fluid released near or above the supercritical point, resulting in a high velocity gas-vapor flow. The high velocity gas flow effectively reduces the gas diffusion boundary layer in front of the deposition surface, thereby improving film quality and deposition efficiency. When the air velocity is higher than the flame velocity, a pilot light or other ignition device must be used to establish a steady state flame. In some cases, two or more pilot lights are required to ensure complete combustion. Using a plasma torch, high speeds are readily obtained without the need for a pilot lamp, following operating conditions known to those of ordinary skill in the art.

Fluids containing solutes do not necessarily serve as fuels for combustion. A nonflammable fluid such as water, _N2O , or _CO2 , or a difficult-to-combust fluid such as ammonia, can be used to dissolve the precursor, or can be used as the second solution compound. They then expand into a flame or plasma torch that provides an environment for the precursors to react. Deposition can be performed at ambient pressure or above or below. Plasma torches work well under reduced pressure. The flame can be stabilized down to 10 Torr and works well at high pressure. Cool flames even below 500°C can be formed at lower pressures. While both can be worked in an open atmosphere, it is preferred to practice the process of the invention in a reaction chamber under a controlled atmosphere to prevent entrainment of airborne impurities in the resulting coating. Many electrocoating and optical coating applications require the absence of these impurities in the coating. These applications generally require thin films, but thicker films can also be deposited for thermal insulation, corrosion and wear applications.

By extending the deposition time even further, larger bulk materials (including single crystals) can be grown. Higher deposition temperatures result in faster epitaxial deposition rates due to higher diffusion rates, which are necessary for depositing single crystal thick films or bulk materials.

CCVD is a flame process that utilizes oxygen. Although CCVD can be used to deposit oxygen-reactive materials (oxygen-reactive materials) in the reducing flame part of the flame, the better technology for depositing oxygen-reactive materials (such as nickel) is 1998 4 Related methods described in U.S. Patent Application No. 09/067,975, filed on March 29, the contents of which are incorporated herein by reference.

The invention described in referenced U.S. Patent Application No. 09/067,975 provides an apparatus and method for chemical vapor deposition in which the atmosphere in the coating deposition zone is established by carefully controlling and protecting the materials added to form the coating , and the gas discharged from the deposition zone passes through a barrier zone where the gas exits the deposition zone at an average velocity greater than 50 ft/min (preferably greater than 100 ft/min). flow in the area. The rapid flow of gas through the isolation zone substantially prevents the migration of gases from the ambient atmosphere into the deposition zone where they may react with the coating or the materials forming the coating. By incorporating the various paint precursors in fixed proportions in a liquid medium, careful control can be exercised over the materials used to form the coating. The liquid medium is atomized as it is introduced into the reaction zone where the liquid medium is vaporized and the paint precursors react to form reacted paint precursors. In addition, the paint precursor can be added in gaseous form (as a gas itself or as a mixture in a carrier gas). The reacted paint precursor, typically consisting of partially, totally and fractionally reacted components, can flow as a plasma into the deposition zone. The reacted paint precursor contacts the substrate surface in the deposition zone and deposits a coating thereon. A flowing curtain of inert gas may be provided around the reaction zone to protect the reactive coating/plasma in the zone from contamination by materials used in surrounding instruments or components of the ambient atmosphere.

Both the vaporization of the liquid medium and the reaction of the paint precursors in the reaction zone require energy input. The required energy can be provided by a variety of energy sources such as resistive heating, induction heating, microwave heating, RF heating, hot surface heating and/or mixing with hot inert gases.

The non-combustion process is referred to herein as "controlled atmosphere combustion chemical vapor deposition" (CACCVD). This technique provides a relatively controlled rate of energy input, enabling high rates of coating deposition. In some preferred examples, the liquid medium and/or the secondary gas used to atomize the liquid medium may be a combustible fuel used in CACCVD. Of particular importance is the ability of CACCVD to form high quality adherent deposits at or near atmospheric pressure, thereby eliminating the need for complex vacuum or similar isolation equipment. For these reasons, in many cases CACCVD thin film coatings can be applied in situ or "in situ" where the substrate is.

Combustion chemical vapor deposition (CCVD) is not suitable for those coating applications that require an oxygen-free environment. Suitable for these applications is CACCVD, which uses non-combustible energy sources such as hot gases, heated tubes, radiant energy, microwaves, and photons excited using infrared or laser sources. In these applications it is important that all liquids and gases used are free of oxygen. The paint precursor can be added as a solution or suspension in a liquid, such as ammonia or propane, suitable for depositing nitrides or carbides, respectively.

The CACCVD process and equipment control the atmosphere in the deposition area, thereby enabling the production of sensitive coatings on temperature-sensitive or vacuum-sensitive substrates that can be larger than those processed by traditional vacuum chamber deposition techniques substrate.

Another advantage of CACCVD is its ability to coat substrates without providing additional energy to the substrate. Thus, this system can be used to coat substrates that were previously intolerant of the temperatures encountered with most existing systems. For example, a nickel coating can be applied to a polyimide sheet substrate without causing deformation of the substrate. Existing atmospheric pressure deposition techniques cannot carry out chemical vapor deposition of metallic nickel because nickel has a strong affinity for oxygen, and vacuum treatment of polyimide sheet substrates is also problematic because it is exposed to heat. It will dehydrate under vacuum and is prone to dimensional instability.

Description of drawings

Figure 1 is a schematic diagram of the apparatus of the present invention.

Figure 2 is a schematic diagram of an apparatus for depositing films and powders using near supercritical and supercritical atomization.

Figures 3a-3c are detailed schematic diagrams of nebulizers used in the present invention.

4a-4c are cross-sectional views illustrating steps of forming a sheet resistance according to the present invention. Figure 4d is a plan view of the sheet resistance of Figure 4c.

5a-5c are similar cross-sectional views illustrating the steps of forming a sheet resistor according to another method of the present invention.

Figure 6 is a cross-sectional view of the resistor of Figure 4c embedded in an insulating material.

Fig. 7 is a schematic partial sectional view of an apparatus for applying paint according to the present invention.

FIG. 8 is a close-up perspective view, partly in section, of a portion of the applicator head used in the apparatus of FIG. 7. FIG.

Figures 9a-9g are cross-sectional views of a structure showing the process of making a resistor from a freestanding foil coated with a resistive material.

Figures 10a, b and c illustrate a method for starting a resistive pattern on a metal foil with a three-layer laminate comprising a conductive foil layer, an etchable intermediate layer and a layer of porous resistive material.

Contents of the invention

According to the invention, a sheet resistor is formed on a substrate, which resistor can be embedded in a printed wiring board. A thin layer of resistive material is formed on the substrate. A preferred resistive material to form a thin layer is a homogeneous mixture of a metal such as platinum and a dielectric material such as silicon dioxide and aluminum oxide. Even a small amount of dielectric material mixed with a metal can significantly increase the resistance of the metal. Preferably the resistive material is deposited on the substrate by combustion chemical vapor deposition (CCVD). In the case of zero-valent metals and dielectric materials, a homogeneous mixture is obtained by co-depositing metal and dielectric materials by CCVD. In order to form discrete patches of the resistive material (discrete patches of the resistive material), select portions of the resistive material layer are etched away. Thus, a layer of resistive material is covered with a patterned resist (eg, exposed and developed photoresist), and the exposed portions of the underlying layer of resistive material are etched away. In addition, the present invention forms thin layers of discrete pieces of resistive material, and conductive material electrically connecting spaces between the layers of resistive material, said conductive material electrically connecting regions of resistive material with circuits. These structures of insulating material, resistive material and conductive material can be formed by selective etching methods.

Certain resistive materials that can be deposited by CCVD of the present invention are porous. This porosity facilitates etching with etchant that attacks the underlying substrate. Selected portions of the layer of porous resistive material are exposed to an etchant that permeates through the resistive layer through the pores and etches the underlying substrate thereby breaking the bond between the substrate and the layer of resistive material. Since the layer of resistive material is very thin, when the bond is broken, the thin layer of resistive material in the area in contact with the etchant breaks and is carried away in the etchant (such as sprayed etchant). The time of contact with the etchant should be limited to a period of time sufficient to remove (strip) the resistive material but not long enough to cause significant underetching of the substrate.

In one embodiment of the invention, the resistive material is deposited on a metal foil, particularly copper foil, which is used to form conductive circuit traces in electrical contact with the sheet resistors of the present invention. Discrete blocks of resistive material are formed using photoimaging and ablative etching. The side of the foil with the layer of resistive material is then embedded in a dielectric material such as prepreg. The foil is then etched into a circuit trace pattern using photoimaging. The circuit trace pattern is also embedded in the dielectric material.

Because copper and/or copper oxide interact with the resistive material layer as it is deposited, in one embodiment of the invention a barrier layer is deposited on the copper surface prior to depositing the resistive material layer. The barrier layer can be a metal such as nickel or a layer of dielectric material such as silicon dioxide and is so thin that it does not interfere with the electrical contact between the copper foil and the resistive material deposited thereon. When the resistive material layer is porous, exfoliation etching can be accomplished using an etchant that corrodes the barrier layer.

The present invention can be understood more readily by reference to the following detailed description of preferred embodiments of the invention and the accompanying drawings.

It should be understood that the technical terms used in the present invention are only used to describe specific embodiments, and do not limit the present invention. It must be noted that as used in this specification and the appended claims, the singular forms "a" and "the" include plural referents unless the context clearly dictates otherwise.

Throughout this application, where publications are cited, the entire contents of those publications are hereby incorporated by reference into this application in order to more fully describe the state of the art to which this invention pertains.

The present invention provides a method of coating a substrate with a selected material. The method comprises: dissolving one or more reagents capable of reacting to form a selected material in a suitable carrier at a first selected temperature and a first selected pressure to form a transport solution, wherein, For a single precursor reagent, precipitation of the reagent from solution or a change in chemical bond is considered a "reaction" herein. Some time before the actual deposition, the substrate is placed in a zone with a second selected pressure. The second selected pressure may be ambient pressure, typically above 20 Torr. The transfer solution is then pressurized to a third selected pressure higher than the second selected pressure with the pressure regulator. Those skilled in the art will recognize that there are many suitable pressure regulating devices including, but not limited to, compressors and the like. Next, the pressurized transport solution is introduced into a fluid conduit having an inlet end and an opposite outlet end, the outlet end being fitted with a temperature regulating device to regulate the temperature of the solution at the outlet end. The outlet end of the conduit also includes an outlet port oriented to direct fluid in the conduit into the area and toward the substrate. The exit orifice can be similar in shape to nozzles or restrictors used in other spray and atomization applications. Then, the solution is heated to a second selected temperature (in the range of 50°C above and below the critical temperature _Tc of the solution) using a temperature regulating device, while using a pressure regulating device to maintain a third selected pressure above the second selected pressure and Above the corresponding liquidus pressure or critical pressure P _c of the solution at the second selected temperature. The pressurized, heated solution is then introduced into the zone through the outlet orifice of the conduit to produce a sprayed solution spray in the direction of the substrate. As the solution is introduced into the area, one or more selected gases are mixed into the sprayed solution spray to form a reactive spray, after which the reactable spray is exposed to an energy source at selected energization points Down. The energy source provides sufficient energy to react the reactive spray liquid (containing one or more reagents of the delivery solution) to form the substrate-coated material and coating.

In another embodiment of the invention, the energy source includes a flame source and the selected energization point includes an ignition point. In another embodiment, the energy source includes a plasma torch.

In another embodiment of the method, the second selected pressure of the zone is ambient pressure.

In another embodiment, the sprayed solution spray is a vapor or aerosol having a maximum droplet size of less than 2 microns.

In another embodiment, the second selected pressure of the zone is reduced to produce a combustion flame with a temperature below 1000°C.

In another embodiment, the carrier is propane and the transport solution comprises at least 50% propane by volume. In another embodiment, the transport solution further includes butanol, methanol, isopropanol, toluene, or combinations thereof. In another embodiment, the carrier is selected such that the transport solution is substantially free of precipitates at standard temperature and pressure for a period of time sufficient to carry out the method.

In another embodiment of the method, a pressurized vessel is used to bring a substance that is a gas at standard temperature and pressure to a selected pressure sufficient to form a liquid or a supercritical fluid before, during or after the pressurization step. under (depending on temperature) contact with the transfer solution. In a preferred embodiment, the transport solution comprising a gas at a standard temperature and pressure is substantially free of precipitates at the selected pressure for a period of time sufficient to carry out the process. In another embodiment, the reagent concentration of the transport solution is between 0.0005M and 0.05M.

In another embodiment, the outlet end of the conduit further includes a fluid introduction port through which fluid is added to the pressurized, heated solution prior to directing the pressurized, heated solution through the outlet port of the conduit. This introduction forms a mixed solution with a lowered supercritical temperature.

In another embodiment, the vapor pressure of each of the one or more reagents is not less than about 25% of the vapor pressure of the carrier.

In another embodiment, the outlet end of the conduit comprises a conduit having an inner diameter of 2-1000 microns, more preferably 10-250 microns. In a more preferred embodiment, the outlet end of the conduit comprises a conduit having an inner diameter of 25-125 microns. In another preferred embodiment, the outlet end of the conduit comprises a conduit having an internal diameter of 50-100 microns.

In another embodiment, the temperature regulating means comprises means for resistively heating the conduit by applying a current to the conduit at a selected voltage with a current source. In a preferred embodiment, the voltage is less than 115 volts. In another embodiment, the means for resistively heating the catheter includes a contactor located within 4 millimeters of the exit orifice.

Furthermore, the present invention also provides such above methods, wherein the carrier and the one or more reagents are selected such that the second selected temperature is ambient temperature.

The methods described above may be practiced where the material coating the substrate comprises a metal, an oxide of a metal or metalloid, or a mixture of a metal and an oxide of a metal or metalloid.

In another embodiment, the reactive jetting fluid comprises a combustible jetting fluid having a combustible jetting fluid velocity greater than the flame velocity of the flame source at the point of ignition, and one or more ignition aids to ignite the combustible jetting fluid. spray liquid. In a preferred embodiment, each of the one or more ignition aids comprises a pilot lamp. In another embodiment, the flammable jet velocity is greater than Mach one.

In another embodiment, the ignition point or flame front remains within 2 cm of the exit orifice.

The present invention also provides a method wherein, in the exposing step, the substrate is cooled with a substrate cooling device. In a preferred embodiment, the substrate cooling means includes means for directing water onto the substrate. However, those of ordinary skill in the art will recognize that many other suitable cooling arrangements may also be used.

In another embodiment, the thickness of the material coating the substrate is less than 100 nanometers. In another embodiment, the material coating the substrate comprises a graded composition. In another embodiment, the material of the coating material comprises an amorphous material. In another embodiment, the material coating the substrate comprises a nitride, carbide, boride, metal, or other oxygen-free material.

The present invention also provides a method that also includes flowing a selected sheath gas around the reactive spray liquid to reduce entrained impurities and maintain a favorable deposition environment.

In a preferred embodiment, the second selected pressure is greater than 20 Torr.

Referring now to FIG. 1 , a preferred apparatus 100 includes a pressure regulating device 110 (such as a pump) for pressurizing a transfer solution T (also referred to as "precursor solution") in a transfer solution reservoir 112 to a first selected pressure, wherein the transfer solution T comprises a suitable carrier in which one or more reagents capable of reacting to form the selected material are dissolved, wherein the pressurizing means 110 is capable of maintaining the first selected pressure at the transfer solution T at the corresponding liquidus pressure at its temperature (if the temperature is below _Tc ) or above the critical pressure _Pc ; Fluidly connected, the outlet end 124 has an outlet hole 126 oriented to direct the fluid in the conduit 120 into a second selected pressure zone 130 lower than the first selected pressure, and toward the direction of the substrate 140 direction. Wherein outlet hole 126 also comprises the device 128 (referring to Fig. 2 and 3, atomizer 4) that is used to spray solution to form sprayed solution spray liquid N; The place where the end 124 is thermally connected is used to adjust the temperature of the solution at the outlet end 124 within the range of 50°C above and below the supercritical temperature _Tc of the solution; the gas supply device 160 is used to mix a liquid N into the sprayed solution. One or more gases (such as oxygen) (not shown) to form a reactive spray liquid; energy source 170, located at a selected energization point 172, is used to react the reactive spray liquid, whereby the energy source 170 Sufficient energy is provided to react the reactable spray liquid in the second selected pressure zone 130 to coat the substrate 140 .

In another embodiment of the apparatus, the energy source 170 includes a flame source and the selected energization point 172 includes an ignition point. In another embodiment, the energy source 170 includes a plasma torch. In another embodiment, outlet port 126 also includes a pressure restrictor (see FIG. 3, restrictor 7).

In another embodiment of the apparatus, the second selected pressure of the zone is ambient pressure.

In another embodiment, the sprayed solution spray N is a vapor or an aerosol having a maximum droplet size of less than 2 microns.

In another embodiment, the second selected pressure of the zone is reduced to produce a combustion flame with a temperature below 1000°C.

In another embodiment, the carrier is propane and the transport solution comprises at least 50% propane by volume. In another embodiment, the transport solution further includes butanol, methanol, isopropanol, toluene, or combinations thereof. In another embodiment, the carrier is selected such that the transport solution is substantially free of precipitates at standard temperature and pressure for a period of time sufficient to carry out the method.

In another embodiment of the apparatus, a pressurized vessel (not shown) is provided, and a substance that is a gas at standard temperature and pressure is mixed with the transfer solution at a pressure selected to be sufficient to form a liquid or supercritical fluid. touch. In a preferred embodiment, the transport solution comprising a gas at a standard temperature and pressure is substantially free of precipitates at the selected pressure for a period of time sufficient to carry out the process. In another embodiment, the reagent concentration of the transport solution is between 0.0005M and 0.05M.

In another embodiment, the outlet end 124 of the conduit 120 also includes a fluid introduction hole (see FIG. Fluid is added to the pressurized, heated solution through the fluid introduction port. This introduction forms a mixed solution with a lowered supercritical temperature.

In another embodiment, the vapor pressure of each of the one or more reagents is not less than about 25% of the vapor pressure of the carrier.

In another embodiment, the outlet end of the conduit comprises a conduit having an inner diameter of 2-1000 microns, more preferably 10-250 microns. In a more preferred embodiment, the outlet end of the conduit comprises a conduit having an inner diameter of 25-125 microns. In another preferred embodiment, the outlet end of the conduit comprises a conduit having an internal diameter of 50-100 microns.

In another embodiment, the temperature regulating means 150 includes means for resistively heating the conduit by applying a current to the conduit with a current source at a selected voltage. In a preferred embodiment, the voltage is less than 115 volts. In another embodiment, the means for resistively heating the catheter includes a contactor 152 located within 4 millimeters of the exit orifice (126).

Furthermore, use of the above-described apparatus is conditioned on selecting the carrier and one or more reagents such that the second selected temperature is ambient temperature.

The apparatus described above may be used where the material coating the substrate 140 includes metal. Alternatively, the material coating substrate 140 includes one or more metal oxides. In another embodiment, the material coating substrate 140 includes at least 90% silica.

In another embodiment, the reactive jet comprises a combustible jet having a combustible jet velocity greater than the flame velocity at the flame source ignition point 172, and one or more ignition aids 180 to Ignite flammable spray fluid. In a preferred embodiment, each of the one or more ignition aids 180 includes a pilot light. In another embodiment, the combustible jet velocity is greater than Mach 1.

In another embodiment, the ignition point 172 or flame front remains within 2 cm of the exit aperture.

The present invention also provides a substrate cooling device 190 for cooling the substrate 140 . In a preferred embodiment, substrate cooling means 190 includes means for directing water onto substrate 140 . However, those of ordinary skill in the art will recognize that many other suitable cooling arrangements may also be used.

In another embodiment, the thickness of the material coating substrate 140 is less than 100 nanometers. In another embodiment, the material coating substrate 140 comprises a graded composition.

Also provided is an apparatus comprising a device (see Figures 2 and 3, feed lines 17 and 19) for flowing a selected outer shielding gas around the reactive spray liquid to reduce entrainment impurities and maintain a favorable deposition environment.

In a preferred embodiment, the second selected pressure is greater than 20 Torr.

In another embodiment of the method, the energy source includes a flame source and the selected energization point includes an ignition point. In another embodiment, the energy source includes a plasma torch, hot gas, or the like.

In another preferred embodiment of the powder forming method, the delivery solution has a concentration between 0.005M and 5M.

To simplify operation, it is useful to pump the precursor/solvent solution into the nebulizer at room temperature. Heating the solution should be done as the last step immediately before releasing the solution into the lower pressure zone. This post-heating minimizes the reactions and immiscibility that occur at higher temperatures. Keeping the solution below the supercritical temperature until atomization keeps the amount of dissolved precursor within the normal solubility region and reduces the possibility of significant solvent-precursor concentration gradients forming in the solution. These solubility gradients are due to the sensitivity of the solution concentration of the supercritical solvent to pressure. It was found that small pressure gradients (as they can develop along the precursor-solvent system transport) lead to significant changes in solubility. For example, the solubility of acridine in carbon dioxide at 308°K can be increased by a factor of 1000 by increasing the pressure from 75 to 85 atm. See "Supercritical Fluid Nucleation of Difficult to Comminute Solids" by V. Krukonis published at the AIChE Conference held in San Francisco on November 25-30, 1984. Such solubility changes are potentially detrimental as they may allow precursors to fall out of solution and precipitate or react prematurely, clogging lines and filters.

The sudden pressure drop and high velocity at the nozzle cause the solution to expand and atomize. For solute concentrations in the normal solubility range (which is the preferred mode of operation of the near supercritical atomization system of the present invention), the precursors are effectively still in solution after the solution is sprayed into the low pressure region. The term "effectively in solution" must be understood in conjunction with the processes that occur when a solution with a higher solute concentration than normal solvent concentration is injected into a region of low pressure. In this case, the pressure drop makes the high supersaturation the cause of severe solute nucleation. If severe nucleation rapidly depletes the solvent from all dissolved precursors, then the proliferation of small precursor particles increases. See "Rapid Expansion of Supercritical Fluid Solutions: Solute Rormation of Powders, Thin Films, and Fibers" by D.W. Matson, J.L. Fulton, R.C. Petersen, and R.D. Smith ("Rapid Expansion of Supercritical Fluid Solutions: Solute Formation of Powders, Films, and Fibers") "), Ind.Eng.Chem.Res., 26, 2298 (1987); "Vapor Phase Processing of Powders: Plasma Synthesis and Aerosol Decomposition" by H.Anderson, T.T.Kodas and D.M.Smith ("Vapor Phase Processing of Powders: Plasma Synthesis Body synthesis and aerosol decomposition"), Am.Ceram.Soc.Bull., 68, 996(1989); C.J Chang and A.D Randolph, "Precipitation of Microsize Organic Particle from Supercritical Fluid" ("Precipitation of Microsize Organic Particle from Supercritical Fluid" Organic Particles"), AIChE Journal, 35, 1876 (1989); "Generation of Complex MetalOxides by aerosol Processes: Superconducting Ceramic Particles and Films" by T.T. Kodas ("Generation of Complex MetalOxides by aerosol Processes: Superconducting Ceramic Particles and Films" Particles and membranes"), Adv.Mater., 6, 180(1989); E.Matijevic's "Fine Particles: Science and Technology" ("fine particles: Science and Technology"), MRS Bulletin, 14, 18(1989) ; E.Matijevic, "Fine Particles Part II: Formation Mechanisms and Applications" ("Fine Particles Part II: Formation Mechanism and Applications"), MRS Bulletin, 15, 16 (1990); K.P.Johnston and J.M.L.Penniger edited "Supercritical Fluid "Solid Formation After Expansion of Supercritical Mixtures" by R.S.Mohamed, D.S.Haverson, P.G.Debenedetti and R.K.Prud'homme, p. 355 of "Science and Technology" ("Supercritical Fluid Science and Technology") solid formation from supercritical mixtures"), American Chemical Society, Washington, DC (1989); "Effects of Process Conditions on Crystals Obtained from Supercritical Mixtures" by R.S.Mohamed, P.G.Debenedetti and R.K.Prud'homme ("Processing Conditions on Crystals Obtained from Supercritical Mixtures" Effect of crystals"), AIChE J., 35, 325 (1989); "Formation of Bioerodible Polymeric Microspheres and Microparticles by Rapid Expansion of Supercritical Solutions" by J.W.Tom and P.G.Debenedetti ("Formation of Bioerodible Polymeric Microspheres and Microparticles by Rapid Expansion of Supercritical Solutions" Polymer Microspheres and Microparticles"), Biotechnol. Prog. 7, 403 (1991). Particles are not suitable for forming thin coatings, but may be beneficial in forming powders.

Thus, compared to non-heated devices that rely on rapid expansion of the solvent, all operating above the supercritical temperature without exception, heated atomizers offer the additional advantage of (1) the temperature used to nebulize the precursor-solvent mixture. The degree of atomization is well controlled, and (2) severe nucleation of the precursors can be avoided while still having the advantages of supercritical atomization. Supersonic velocities can be generated to form a mach disk, which additionally facilitates atomization. Adding gas to the released atomized material helps direct the flow and ensures the desired combustion mixture.

Liquid solutions can be vaporized to varying degrees by adjusting the heat input into the atomizing device. Liquid solutions of higher supercritical temperature liquids (which are liquid at STP) can emerge as streams significantly away from the supercritical state without inputting heat into the atomizing device. This would result in poor flame formation and possibly unwanted contact of the liquid with the substrate. Reducing the temperature difference between the liquid solution at the nozzle and its supercritical temperature enables the liquid solution to break down into droplets, which form a mist and are released from the atomizing device. After a short distance these droplets vaporize and thus become invisible. As the atomization device approaches the supercritical temperature, the size of the liquid solution droplets becomes smaller and the distance for the solution to vaporize becomes shorter. Using this nebulizer, the vapor droplet size was measured with an aerosol vaporization tester, and the resulting droplet size was below the instrument detection limit of 1.8 microns.

Further increasing the heat input produces a spray-free state at the tip, ie complete vaporization. Without wishing to be bound by theory, this behavior of the solution may be due to the overall supercritical properties of the reagents and solvents. Solutions of precursors in solvents with lower supercritical temperatures (they are gases at STP) behave similarly, but solutions ejected from a pointed nozzle (also called a "nozzle" or "restrictor") even without Heat input also does not create liquid flow. The amount of heat required to obtain optimal solution vaporization is primarily dependent on the heat capacity of the solution and the temperature difference between the supercritical temperature of the solvent and the ambient temperature near the nozzle.

It is desirable to maintain the pressure and temperature of the system (before vaporization) above the boiling point and supercritical point of the solution. If the pressure drops below the liquidus pressure or the critical pressure, while the temperature is above the boiling point, then vaporization of the solvent occurs in the pipe before the nozzle. This leaves solutes behind which can accumulate and clog the nebulizer. Similarly, the pressure is preferably high enough in the supercritical region to make the fluid more like a liquid. Liquid-like supercritical fluids are better solvents than more gas-like supercritical fluids, further reducing the likelihood of solutes clogging the nebulizer. If the precursor-to-precursor interaction is higher than the strength between the solvent and the precursor, then the solvent-precursor bond breaks, effectively pulling the precursor out of solution. Precursor molecules can then form agglomerates, stick to the nebulizer and clog the restrictor. In most cases this problem can be solved by moving the vaporization point from the inside of the tip to the end of the tip by reducing the heat input into the atomizer. Another solution is to use a solvent that forms stronger bonds with the precursor, resulting in a more stable solution. A small amount of spray at the tip usually produces the best quality film. If the temperature of the solution is too high or too low, nano or micro sized pellets of material will form. These pellets are detrimental if a dense coating is desired.

If a spray-free state is achieved, the deposition proceeds above the critical temperature. Flame heating and mixing with external gas keeps STP liquid solvents from condensation and droplet formation. In the case of no spray, the atomization and intermixing are very good, but the flow stability is reduced, resulting in flames that run from side to side relative to the direction of the tip. In the case of this flame behavior, although deposition is still possible, it is difficult to form a film with strict thickness uniformity. Furthermore, it is necessary to keep the temperature of the solution prior to release below the temperature at which the solute precipitates or reacts and precipitates. When solvent mixtures are used, the metastable homogeneous limit line of immiscibility may be crossed during heating. This results in the creation of two separate phases, in which the concentration may be different due to the different solubility of the solute. This may affect the formation of precursor and product pellets at high atomization temperatures. All of these factors suggest that it is advantageous to have the solution heated as little as possible up to the tip, if necessary, so that there is not enough time for a potentially unwanted equilibrium state of matter to occur. The structure of the deposited film can thus be precisely controlled.

Due to this control, it is possible to generate a wide variety of membrane microstructures. By increasing the solution concentration, it is possible to increase the deposition rate and the following microstructure changes with the increase of the solution concentration: from dense to porous, from metallic luster to dull, from smooth to rough, from columnar to hillock shape, and from thin to thick. Gradient and multilayer coatings can also be produced. Multiple layers can be formed by feeding solutions containing different precursors to a single flame. Sequential multilayer deposition flames can be used to increase the throughput of manufacturing applications. Other factors that control deposition parameters include: substrate surface temperature that controls surface diffusion and nucleation; pressure that controls boundary layer thickness and thus deposition rate; layer growth habit; flame and plasma energy levels affect where reactions occur and vapor stability; and distance from substrate affects time from spray to reaction to deposition, which can enable particle formation or increase diffusion time to form Larger pellets. In addition, electric and magnetic fields can affect the growth behavior of some materials, or increase deposition efficiency. Those of ordinary skill in the art will recognize that such electric and magnetic fields can affect the growth behavior of some vapor deposited materials and alter the specific deposition rate and efficiency.

Because the energy input required for a solution-heated atomizer will vary for different precursor/first solvent/second solvent solutions, it is preferable to use a solution with a constant ratio of first solvent to second solvent Deposit multiple layers of thin film. This way, it is not necessary to change the energy input to the nebulizer when switching from one solution to another. This simplification of the scheme improves performance and reliability while reducing cost. In addition, the substrate can be subjected to flames containing different reagents to produce the desired multilayers.

As the solution provides the fuel for combustion, concentrations up to 0.1 molar give dense coatings, depending on the material. Preferred concentrations are up to 0.01 molar for most materials. Materials with poor diffusivity and mobility require solution concentrations below 0.002. Solution concentrations below 0.0001 molar give very slow deposition rates for most materials. Flame deposition with added combustible materials can have higher concentrations, even over 1M, but for better vaporization of the precursors, high concentrations are not desirable unless the precursor or precursors have a high vapor pressure. The solution concentration of the low vapor pressure precursor is preferably less than 0.002 molar.

Without wishing to be bound by theory, it is helpful to understand that the principle of the deposition technique of the present invention is based on the discovery that CVD is not limited to reactions on surfaces. See "Combustion Chemical Vapor Deposition, a Novel Thin Film Deposition Technique" by Hunt, A.T. ("Combustion Chemical Vapor Deposition, a New Thin Film Deposition Technique"), Ph.D. Thesis Georgia Inst.of Tech, Atlanta, GA., ( 1993); Hunt, A.T., "Presubstrate Reaction CVD, and a Definition for Vapor" ("Presubstrate Reaction CVD and Vapor Definition") presented at the 13th International Conference on CVD, Los Angles, CA (1996), these contents References are incorporated herein. The reaction can take place primarily in the gas stream, but the resulting deposited material must be subcritical in size to obtain a coating with a vapor-deposited microstructure. These findings suggest that vapors consist of individual atoms, molecules, or nanoaggregates that can adsorb on substrates and readily diffuse into low-energy sites or structures. Therefore, the maximum aggregate size must decrease with decreasing substrate temperature because the critical core size also decreases with decreasing temperature. It is known to those of ordinary skill in the art that reagent agglomerates remain after solvent evaporation, the size of the agglomerates being related to the vapor pressure of the reagent, the initial droplet size and the concentration of the solution. Therefore, atomization of low vapor pressure reagents (and thus not vaporizing in the flame) must be very fine to form the vapor.

Preferred liquid solvents are low cost solvents including but not limited to ethanol, methanol, water, isopropanol and toluene. Aqueous solutions must be added to a pre-existing flame, whereas flammable solvents themselves can be used to form a flame. It is preferred, but not necessary, to use the solution to form the body of the flame rather than add the solution to the flame. Lower reagent concentrations give results that favor the formation of material below the critical core size.

A preferred solvent and second solution fluid is propane, which is a gas at STP. It must be noted, however, that many other solvent systems may also be used. See, eg, CRC Handbook of Chemistry and Physics , CRC Press, Boca Raton, Florida. Propane is preferred because of its low cost, commercial availability and safety. A number of low-cost organometallic compounds are available for primarily propane-containing solutions. For ease of handling, initial precursors can be dissolved in methanol, isopropanol, toluene, or other propane-compatible solvents. This initial solution is then placed in a container to which liquid propane is added. Propane is liquid only above about 100 psi at room temperature. The supercritical point of the resulting solution is much lower than that of the initial solution, which reduces the energy input required by the nebulizer and facilitates nebulization. In addition, the function of the primary solvent is to increase the polar solubility of propane so that, for many reagents, its solution concentration is higher than that obtained with propane alone. As a general rule, the polarity of the first solvent should increase as the polarity of the solute (precursor) increases. Therefore, isopropanol contributes more to the solubility of polar solutes than toluene. In some cases, the first solvent acts as a barrier between the second solvent and the ligand on the solute. An example is platinum acetylacetonate [Pt(CH ₃ COCHCOCH ₃ ) ₂ ] dissolved in propane, where the weight ratio between precursor/first solvent and first solvent/second solvent can be higher than required for other systems The weight ratio.

Ammonia was considered and tested as a secondary solvent for depositing coatings and powders. While ammonia is an inexpensive solvent that is compatible with some nitrate-based precursors, it is not easy to use with other secondary solvents, and the general corrosivity of pure ammonia creates problems. The atomization performance of ammonia was tested without the addition of precursors, and the pressure vessels used were significantly corroded after the tests, even with inert type 316 stainless steel vessels. In contrast to hydrocarbon-based solvents, ammonia renders Buna-N and Viton gaskets useless after only a few minutes. Even with suitable gaskets, problems can arise because the required coating or powder must generally not contain traces of iron or other elements that leach from the pressure vessel walls. However, materials such as EPDM elastomers may be used.

Other gas-like second solvents that have been tested and may be used include ethane, ethylene, ethane/ethylene mixtures, propane/ethylene mixtures and propane/ethane mixtures. Platinum thin films were deposited from supercritical mixtures of ethane and platinum metal-organic compounds.

A useful solvent and second solution fluid is propane, which is a gas at STP. It must be noted, however, that many other solvent systems may also be used. See, eg, CRC Handbook of Chemistry and Physics , CRC Press, Boca Raton, Florida. Propane is preferred because of its low cost, commercial availability and safety. A number of low-cost organometallic compounds are available for primarily propane-containing solutions. For ease of handling, initial precursors can be dissolved in methanol, isopropanol, toluene, or other propane-compatible solvents. This initial solution is then placed in a container to which liquid propane is added. Propane is liquid only above about 100 psi at room temperature. The supercritical point of the resulting solution is much lower than that of the initial solution, which reduces the energy input required by the nebulizer and facilitates nebulization. In addition, the function of the primary solvent is to increase the polar solubility of propane so that, for many reagents, its solution concentration is higher than that obtained with propane alone. As a general rule, the polarity of the first solvent should increase as the polarity of the solute (precursor) increases. Therefore, isopropanol contributes more to the solubility of polar solutes than toluene. In some cases, the first solvent acts as a barrier between the second solvent and the ligand on the solute. An example is platinum acetylacetonate [Pt(CH ₃ COCHCOCH ₃ ) ₂ ] dissolved in propane, where the weight ratio between precursor/first solvent and first solvent/second solvent can be higher than required for other systems The weight ratio.

Ammonia was considered and tested as a secondary solvent for depositing coatings and powders. While ammonia is an inexpensive solvent that is compatible with some nitrate-based precursors, it is not easy to use with other secondary solvents, and the general corrosivity of pure ammonia creates problems. The atomization performance of ammonia was tested without the addition of precursors, and the pressure vessel used was significantly corroded after the test, even with an inert type 316 stainless steel vessel. In contrast to hydrocarbon-based solvents, ammonia renders Buna-N and Viton gaskets useless after only a few minutes. Even with suitable gaskets, problems can arise because the required coating or powder must generally not contain traces of iron or other elements that leach from the pressure vessel walls. However, materials such as EPDM elastomers may be used.

Other solvents and solvent mixtures tested gave similar quality, but were more difficult to work with because of their significantly lower boiling points, which required cooling of the solution or very high pressure. Propane is the preferred solvent due to its ease of handling, but other supercritical solvents may be used as propane substitutes in cases where propane cannot be used (eg, when no propane-soluble precursors can be found). Other fluids can be used to further lower the supercritical temperature if desired.

One method of heating is to apply an electrical current between the end of the nozzle (where the precursor solution is sprayed into the low pressure region) and the rear of the restrictor tube. This method of directly heating the restrictor tube facilitates rapid changes in atomization conditions due to its short response time. The location of the strongest heating can be moved towards the tip by increasing the connection resistance between the tip and the wire connecting the tip. Thin-walled restrictors have greater resistance than thick-walled restrictors, resulting in shorter response times. Other heating methods can be applied and several have been investigated including, but not limited to, remote resistive heating, pilot flame heating, induction heating, and laser heating. Other suitable heating means for regulating the temperature at the exit orifice of the atomizer can be readily determined by those of ordinary skill in the art.

Remote resistance heating uses a non-conductive restrictor tube mounted inside the electric heating tube. The non-conductive tube fits snugly inside the conductive tube. Applying an electric current to the conductive tube heats the tube, and the energy is transferred into the non-conductive current-limiting tube inside. Compared with the method of directly heating the restrictor tube, this method requires a higher heating current and exhibits a longer response time, which is advantageous in some situations, because the increased response time can obtain a high degree of thermal stability. On the other hand, long-burning flame heating and laser heating use the energy of long-burning flame and laser, respectively, to heat the restrictor tube. The heating of the restrictor tube can be carried out by direct heating of the nozzle of the restrictor tube in a long-burning flame or laser, or by indirect heating of the larger outer tube. Since the energy required to be transferred into the solution is considerable, the tubes to be heated preferably have thicker walls than in the case of direct electric heating or remote electric heating. Thin-wall restrictors are permitted when the outer tube is in contact with pilot flames or lasers.

Referring now to FIGS. 2 and 3 , an apparatus 200 for depositing films and powders using supercritical atomization is shown. The apparatus 200 comprises a fixed or variable speed pump 1 which pumps a reagent delivery solution 2 (also called "precursor solution") from a solution container 3 into a nebulizer (also called a "nebulizer" or "vaporizer"). )4. FIG. 3 is an illustration of the atomizer 4 in more detail. Precursor solution 2 is pumped from precursor solution container 3 through line 5 and filter 6 into nebulizer 4 . The precursor solution 2 is then pumped into a controlled flow restrictor 7 of constant or variable temperature. Heating can be accomplished in a number of ways including, but not limited to, resistive electric heating, laser heating, induction heating, or flame heating. For resistive electric heating, either AC or DC current can be used. One of the connections 8 to the restrictor 7 is preferably placed very close to the nozzle of the restrictor 7 . In the case of heating by a direct current source, this connection 8 or pole can be positive or negative. The other pole 9 can be connected anywhere else along the current limiter 7 , inside or outside the housing 10 . For special applications (such as coating the inside of pipes), in which case it is advantageous that the atomizer has a small overall volume, it is preferably connected to the flow restrictor 7 at the rear of the housing 10 or inside the housing 10 . The gas connections at the rear of the housing 10 are shown in an arrangement of lines, but could also be arranged in any other way that does not affect the function of the device 200 .

A thin gas A supply line 11, 1/16" ID in most cases, carries the combustible gas mixture to a small outlet 12 where it can be used as a stable long-burning flame to burn the supply through the restrictor 7 Precursor solution, preferably within 2.5 cm of flow restrictor 7. The supply of gas A is monitored by flow controller 13, and the flow of each component 14 and 15 of the gas A mixture is controlled. The fuel component 14 and the oxidizing component 15 of gas A are mixed near or in a mixing "T" junction 16 in the atomizer 4. This post-mixing is desirable for safety reasons because it reduces The possibility of burning back is eliminated.The distribution channel inside the shell 10 connects the gas supply line 11 with the gas A feed port 17.The gas B supply line 18 is used to deliver the gas B from the supply 19, so that it can be mixed with the spray solution Good mixing of the spray liquid. In most cases utilize high velocity air flow. Some gas B supply holes 20 (6 in most cases, more or less holes can be used, depending on the application) are located at the limit Around the flow device 7, gas B is provided to obtain the required flow pattern. The flow performance of the gas flow B is affected by the following factors: the pressure of the gas B in the gas B reservoir 21, the flow rate measured by the flow controller 13 , the diameter of the pipeline 5 and the number of gas supply holes 20. Alternatively, the gas B can be made to flow through a larger tube coaxial with the restrictor 7 and surrounding it. Once the precursor solution 2 is pumped into the precursor supply line 22, The temperature of the precursor solution is then controlled (in the case of electrical heating) by the current flowing through the current limiter 7 determined by the power supply 23. The heating current can then be adjusted so that a suitable amount of atomization (spray, vaporization) takes place This steady, sustained flame can then ignite the sprayed reactive spray and deposit a powder or film on the substrate 24.

Many different coatings have been deposited using the methods and apparatus described in this invention. While propane was used in most cases as the supercritical second solvent (ie, a small amount of high precursor concentration first solvent mixed with a larger amount of second solvent), other solvents have also been used. Other possible second solvents include, but are not limited to, _N2O , ethylene, ethane, and ammonia.

Those of ordinary skill in the art will recognize that the methods and apparatus of the present invention can coat nearly any substrate. A substrate can be coated if it can withstand the temperature and conditions of the hot gases obtained during processing. The substrate can be cooled using a cooling device (described elsewhere herein), such as a water spray. However, when the substrate surface temperature is low, it is not possible to obtain dense or crystalline coatings of many materials due to the associated low diffusion rates. In addition, substrate stability in hot gases can be further addressed by using low-temperature, low-pressure flames, with or without additional substrate cooling.

A number of chemical precursors have been proposed for CCVD deposited films and powders, and others are proposed herein. In addition to providing the metal or metalloid element, any chemical precursor used for CCVD is required to be soluble in a suitable carrier solvent, preferably propane. Furthermore, if the precursor solution is to contain precursors of more than one metal and/or metalloid, the chemical precursors must be mutually soluble and chemically compatible in a suitable carrier solvent. If the precursor is not highly soluble in the first solvent (such as propane), it can be first dissolved in a second solvent (such as toluene) and subsequently introduced into the first solvent as a solution in the second solvent, provided that the The chemical precursors do not precipitate when the solution is introduced into the first solvent. Furthermore, cost considerations should also be taken into account when selecting chemical precursors.

If and when a mixture of chemical precursors is used to deposit a layer or powder of a particular composition, these precursors can preferably be combined into a homogeneous "pre-solution" without the addition of any additional solvent. Otherwise, all chemical precursors need to be mutually soluble in a common solvent, the less solvent the better, as a "pre-solution". Of course, these required properties facilitate transport and handling, especially when the contemplated primary solvent is propane or other material that is in the gas phase at room temperature. While it is desirable to be able to provide "pre-solutions", it is believed that a deposition solution in which the chemical precursors are miscible in one or more solvents, whether prepared and sold as such or prepared as a deposition solution on-site acceptable.

For deposition, the total concentration of precursor compounds in the carrier solvent is usually between about 0.001-2.5% by weight, preferably between about 0.05-1.0% by weight.

For most CCVD depositions, it is preferred to dissolve the precursors in an organic solvent. Co-precipitation of carbon with the resistive material, however, is undesirable for the resistive materials targeted by the present invention. Some materials, such as nickel, have a high affinity for carbon. Therefore, the precursors of these materials are preferably dissolved in aqueous and/or ammoniacal solutions, in which case the aqueous and/or ammoniacal and/or _N2O solutions will be sucked into the hydrogen/oxygen flame used for CCVD .

One of the advantages of CCVD applied with preferred atomization devices compared to other deposition methods is that the precursor solution containing one or more dissolved chemical precursors is atomized into a near supercritical liquid, or in some In the case of atomization into a supercritical fluid. Thus, the amount of one or more precursors that is combusted and deposited on the substrate or deposited in powder form is independent of the relative vapor pressures of the respective chemical precursors and the one or more carrier solvents. This differs from conventional CVD methods where a separate gas supply line must be provided for each chemical precursor to be vaporized (usually in a carrier gas) to feed the CVD furnace. Furthermore, some conventional CVD precursors disproportionate, which makes it difficult to provide the chemical precursors uniformly, yet another problem that is easily solved by CCVD techniques.

Figures 7 and 8 illustrate controlled atmosphere combustion chemical vapor deposition (CACCVD) equipment. Paint precursor 710 is mixed with liquid medium 712 in formation zone 714 including mixing or reservoir 716 . Precursor 710 and liquid medium 712 form a flowing stream, pressurized by pump 718, filtered by filter 720, added in atomization zone 724 through conduit 722, and flow through reaction zone 726, precipitate accumulation region 728 and isolation region 730. It is not necessary to form a true solution by mixing the prepaint 710 and the liquid medium 712, as long as the prepaint is sufficiently finely divided in the liquid medium. However, it is preferable to form a solution, as this generally results in a more uniform coating.

The flowing stream is atomized as it enters the atomization zone 724 . Atomization can be accomplished by known techniques for atomizing flowing liquid streams. In the illustrated apparatus, atomization is effected by expelling a high velocity atomizing gas immediately around the fluid stream as it exits the conduit 722 . This atomized gas flow is provided by a gas cartridge or other high pressure gas source. In the illustrated embodiment, high pressure hydrogen ( _H2 ) is used as both the atomizing gas and fuel. The atomization gas is passed into the pipeline 738 from the hydrogen cylinder 732 through the regulating valve 734 and the flow meter 736 . Conduit 738 extends coaxially with conduit 722 to the atomization zone where both conduits terminate to allow the high velocity atomizing hydrogen gas to contact the flowing liquid stream, thereby atomizing the liquid stream into a gas/vapor suspended in the surrounding fine particle stream. The stream of fine particles enters reaction zone 726 where the liquid medium evaporates and the precursor paint reacts to form reacted precursor paint, which typically involves dissociation of the precursor paint into its component ions, resulting in a flowing Ionic particles or plasma. The stream/plasma is fed into deposition zone 728 where the reacted paint precursor contacts substrate 740 to deposit a coating thereon.

It is also possible to atomize the flowing stream by injecting an atomizing gas stream directly at the liquid medium/prepaint stream as it emerges from conduit 722 . Alternatively, nebulization can be accomplished by direct application of ultrasonic or similar energy as the fluid stream emerges from conduit 722 .

Evaporation of the liquid medium and reaction of the paint precursors require considerable energy input to the flowing stream before it leaves the reaction zone. This energy input can occur as the stream passes through conduit 722 or in the atomization and/or reaction zones. This energy input can be accomplished by a variety of known heating techniques such as resistive heating, microwave or RF heating, electric induction heating, radiant heating, mixing the flow stream with a remotely heated liquid or gas, photonic heating ( such as lasers). In the illustrated embodiment, energy input is accomplished by combusting fuel and oxidant in direct contact with the flowing stream as it passes through the reaction zone. This relatively new technique, known as Combustion Chemical Vapor Deposition (CCVD), is more fully described in US Patent No. 5,652,021, which is incorporated herein. In the illustrated embodiment, the fuel hydrogen is fed into the pipeline 744 from the hydrogen cylinder 732 through a regulating valve and a flow meter 742 . The oxidant (oxygen) is added into the pipeline 752 from the oxygen cylinder 746 through the regulating valve 748 and the flow meter 750 . Conduit 752 extends coaxially with conduit 744 , which extends coaxially around conduits 722 and 738 . Upon exiting their respective conduits, the hydrogen and oxygen combust to form combustion products which mix with the atomized liquid medium and paint precursor in reaction zone 726, thereby heating and causing the liquid medium to vaporize and react with the paint precursor.

A flowing inert gas barrier surrounding at least the initial portion of the reaction zone isolates the reactive gas from materials present in the apparatus adjacent to the reaction zone. An inert gas such as argon is fed into line 760 from inert gas cartridge 754 through regulating valve 756 and flow meter 758 . Conduit 760 extends coaxially with conduit 752 . Conduit 760 extends beyond the other conduits 722, 738, 744 and 752 to a place close to the substrate where it forms, with substrate 740, deposition zone 728 in which to deposit Coating 762 generally in the cross-sectional shape of pipe 760 . As the inert gas flows out of the end of the oxygen conduit 752, it initially forms a flow barrier extending around the reaction zone, isolating the reactive components in the reaction zone from the conduit 760. As the inert gas travels along conduit 760 , the inert gas mixes with the gas/plasma in the reaction zone and becomes part of the flowing stream entering deposition zone 728 .

Initially igniting the hydrogen and oxygen requires an ignition source. A single manually operated ignition device is sufficient for many applications, however using this device requires temporarily reducing the inert gas flow until a stable flame front is established. In some cases, the entire gas flow may be too large to form an independent stable flame front. In this case, it is necessary to provide an ignition device capable of continuously or semi-continuously igniting the combustible gases as they enter the reaction zone. Examples of ignition sources that may be used are a sustained flame or a spark producing device.

In deposition zone 728 , the reacted paint precursors deposit coating 762 on substrate 740 . The remainder of the flow stream is exhausted from the deposition zone through isolation zone 730 to the environment or surrounding atmosphere. The isolation region 730 serves to prevent contamination of the deposition region by components of the ambient atmosphere. The high velocity of the flow stream as it passes through the isolation zone 730 is characteristic of this zone. In most coating applications, the possibility of contamination of the deposition zone by components of the ambient atmosphere is substantially eliminated by requiring the flow stream to achieve a velocity of at least 50 ft/min through the isolation zone. In those coating operations that are more highly sensitive to fouling (such as making TiN or WC), the possibility of ambient atmosphere fouling the deposition area can be substantially eliminated by requiring the flow stream to achieve a velocity of at least 100 ft/min.

In the embodiment of FIG. 7, a collar 764 is attached to the end of the conduit 760 proximate the deposition zone 728 and extends out perpendicular to the end of the conduit. Isolation region 730 is the space between grommet 764 and substrate 740 . Shaping the guard ring provides a compliant surface 766 close to the surface of the substrate, thereby providing a small gap for the gas to vent from the deposition area to the ambient atmosphere. The gap between the compliant surface 764 of the grommet and the substrate is small enough that for at least a portion of the channel between the grommet and the substrate, the velocity required for the exhaust gas to reach the isolation zone is required. To this end, the compliant surface 764 of the grommet 762 is shaped to lie substantially parallel to the surface of the substrate 740 . In the illustrated embodiment, when the surface of the substrate 740 is substantially planar, the conformable surface of the grommet is also substantially planar.

Edge effects such as elevated temperature and residual reactive species that occur near the end of conduit 760 can extend the deposition zone beyond the region of the substrate directly in front of the end of conduit 760 . Guard ring 764 should extend outwardly from its connection to conduit 760 a distance sufficient to prevent ambient gas from mixing back into the deposition zone due to a possible Venturi effect. And it is ensured that the entire area of the deposition zone (which has been enlarged due to the previously described edge effect) avoids backflow of ambient gas due to the high velocity exhaust blowing through the area between the guard ring and the substrate. The enlarged guard ring ensures protection against contamination throughout the enlarged deposition area. The diameter of the grommet should be at least 2 times the inner diameter of the pipe 760, preferably at least 5 times the inner diameter of the pipe 760. The inner diameter of the conduit 760 is usually in the range of 10-30 mm, preferably in the range of 12-20 mm.

In operation, the grommet 764 is positioned substantially parallel to the surface of the substrate 740 to be coated, at a distance of 1 cm or less from the surface of the substrate. Preferably, the opposing surfaces of the grommet and the substrate are separated by 2-5 mm. Spacers (such as three fixed or adjustable pins, not shown) can be attached to the grommet to help maintain the proper spacing between the grommet and the substrate.

The embodiment shown in Figure 7 is particularly advantageous for applying coatings to substrates that are too large or that cannot be easily handled in a specially controlled environment such as a vacuum chamber or clean room. This illustrated coating technique is advantageous because it can be accomplished under ambient pressure conditions and at a more convenient "on-site" site. A series of coaxial conduits 722, 738, 744, 752 and 760 form an applicator head 768, which can be provided by a smaller flexible hose and is small enough to be portable. Large substrates can be coated in the following ways: one is to make the coating head move laterally along the substrate repeatedly in a grid or similar pattern; Cumulatively provides a uniform coating; one is to raster a row of coating heads. In addition to enabling thin film coatings on articles that were previously too large to be coated, this technique also enables coatings on larger unit substrates that were previously coated under vacuum conditions. Manufacturing economics can be improved by coating these larger units of substrates, especially when large batches of substrates are involved.

The embodiment shown in Figures 7 and 8 is also particularly suitable for producing oxidation-sensitive coatings, such as most metal coatings. To obtain these coatings, fuel is fed through line 744 adjacent to the atomized liquid medium and paint precursor while oxidizer is added through line 752. The atomizing gas added through line 738 and/or the liquid medium added through line 722 may be materials having a combustible value, they may be materials reactive with paint precursors, or they may be inert materials. When the coating or coating precursor material produced is oxygen sensitive, a reducing atmosphere is maintained in the reaction and deposition zone, which can be accomplished as follows: Ensure that the total amount of oxidant added is limited to less than the complete combustion of the fuel supplied to the reaction zone The level of the desired amount, that is to say the oxidizing agent is provided in a substoichiometric amount. Generally, the excess of fuel is limited so as to limit the development of a flame zone as residual hot gases mix with atmospheric oxygen. When coatings and precursor materials are produced that are resistant to oxygen or enhanced by the presence of oxygen (as in the manufacture of most oxide coatings), oxidation can be provided in the reaction and deposition zones by adding stoichiometric or excess oxidizing agents. atmosphere or neutral atmosphere. Additionally, for oxygen resistant reagents and products, oxidizer can be added through inner line 744 and fuel through outer line 752 .

The inert gas provided through conduit 760 must be sufficient to isolate the inner surface of the conduit from the reactive gases produced in the reaction zone. When this inert gas is introduced with other gases from the reaction zone, it must be sufficient to provide the gas velocity required in the isolation zone.

Energy input can be accomplished by mechanisms other than the combustion methods shown in Figures 7 and 8. This can be done, for example, by passing electrical current through the conduit 722 to generate resistive heat in the conduit, which is then transferred to the liquid medium and paint precursor as they pass through the conduit. Obviously, the conduits 722, 738, 744, 752 and 760 are not all required when energy input other than combustion is used. One or both of conduits 744 and 752 are typically omitted when the energy input is by an electrical energy input mechanism.

The porosity or density of the deposited coating can be adjusted by varying the distance between the flame zone and the deposition zone at the substrate surface. In general, reducing this distance increases the density of the coating, while increasing this distance results in a more porous coating.

In the illustrated CACCVD technique, the reaction zone typically expands with the flame created by burning the fuel. Of course, a sufficient distance must be maintained between the flame zone and the substrate so that the substrate is not damaged by the higher temperatures that may result from the flame zone being closer to the substrate surface. Although the sensitivity of the substrate to temperature varies with the substrate material, the temperature in the deposition zone at the substrate surface is usually at least 600°C below the highest flame temperature.

When the energy input is provided by other methods, such as mixing the flow stream with a preheated fluid to provide the main energy input before it is in the reaction zone or before reaching the reaction zone, the maximum temperature obtained in the reaction zone is significantly lower Below the maximum temperature obtained with combustion energy input. In these cases, coating properties can be tuned by varying the distance between the reaction zone and the substrate surface without too much concern for overheating of the substrate. Thus, the terms reaction zone and deposition zone may be used to define functional zones in the apparatus, but are not meant to define mutually exclusive zones, that is, in some cases the reaction of the coating precursors will take place at the surface of the substrate. in the deposition area.

Lower peak temperatures are obtained when the main energy input is by means other than the combustion flame, which enables the use of temperature sensitive coatings (such as some organic materials). Especially in capacitors, integrated circuits or microprocessors, polymers can be deposited as protective coating or dielectric interlayer materials. For example, polyimide coatings can be derived from their polyamic acid precursors. Similarly, polytetrafluoroethylene coatings can be derived from low molecular weight precursors.

The input of energy to the fluent stream before it leaves the reaction zone generally eliminates the need to provide energy to the deposition zone by heating the substrate, as is often required with other coating techniques. In the deposition system of the present invention, since the substrate is used as a heat sink to cool the gases present in the deposition zone, rather than to heat them, the temperature of the substrate is significantly lower than that required to pass through the substrate. The temperature encountered in the system in which the material transmits energy to the deposition zone. Therefore, the CACCVD coating method can be used on many temperature-sensitive substrates that cannot be coated by substrate heating techniques.

A wide range of precursors can be used in the form of gases, vapors or solutions. It is preferred to use the lowest cost precursor that gives the desired morphology. Suitable chemical precursors for depositing various metals or metalloids are as follows, but are not limited to these examples:

Pt Platinum acetylacetonate [Pt(CH ₃ COCHCOCH ₃ ) ₂ ] (in toluene/methanol), platinum-(HFAC ₂ ), diphenyl-(1,5-cyclooctadiene) platinum(II) [Pt (COD) in toluene-propane]platinum nitrate (in aqueous solution of ammonium hydroxide)

Mg Magnesium cyclohexanecarboxylate, magnesium 2-ethylhexanoate [Mg(OOCCH(C ₂ H ₅ )C ₄ H ₉ ) ₂ ], magnesium cyclohexanecarboxylate, Mg-TMHD, Mg-acac, magnesium nitrate, 2 , 4-magnesium glutarate (Mg-2, 4-pentadionate)

Si tetraethoxysilane [Si(OC ₂ H ₅ ) ₄ ], tetramethylsilane, pyrosilicic acid, silicic acid

P Triethyl phosphate [(C ₂ H ₅ O) ₃ PO ₄ ], triethyl phosphite, triphenyl phosphite

La lanthanum 2-ethylhexanoate [La(OOCCH(C ₂ H ₅ )C ₄ H ₉ ) ₃ ], lanthanum nitrate [La(NO ₃ ) ₃ ], La-acac, lanthanum isopropoxide, tris(2, 2,6,6-Tetramethyl-3,5-pimelate)lanthanum [La(C ₁₁ H ₁₉ O ₂ ) ₃ ]

Cr Chromium nitrate [Cr(NO ₃ ) ₃ ], chromium 2-ethylhexanoate [Cr(OOCCH(C ₂ H ₅ )C ₄ H ₉ ) ₃ ], chromium sulfate, chromium carbonyl, chromium(III) acetylacetonate

Ni Nickel nitrate [Ni(NO ₃ ) ₂ ] (in aqueous ammonium hydroxide), nickel acetylacetonate, nickel 2-ethylhexanoate, nickel naphthenol (Ni-napthenol), nickel dicarbonyl

Al aluminum nitrate [Al(NO ₃ ) ₃ ], aluminum acetylacetonate [Al(CH ₃ COCHCOCH ₃ ) ₃ ], triethylaluminum, aluminum sec-butoxide, aluminum isopropoxide, aluminum 2-ethylhexanoate

Pb lead 2-ethylhexanoate [Pb(OOCCH(C ₂ H ₅ )C ₄ H ₉ ) ₂ ], lead naphthenate, Pb-TMHD, lead nitrate

Zr Zirconium 2-ethylhexanoate [Zr(OOCCH(C ₂ H ₅ )C ₄ H ₉ ) ₄ ], zirconium n-butoxide, zirconium (HFAC) ₂ , zirconium acetylacetonate, zirconium n-propoxide, zirconium nitrate

Barium 2-Ethylhexanoate [Ba(OOCCH(C ₂ H ₅ )C ₄ H ₉ ) ₂ ], Barium Nitrate, Barium Acetylacetonate, Ba-TMHD

Nb niobium ethoxide, tetrakis(2,2,6,6,-tetramethyl-3,5-pimelate) niobium

Ti Titanium(IV) isopropoxide [Ti(OCH(CH ₃ ) ₂ ) ₄ ], titanium(IV) acetylacetonate, titanium diacetylacetonate diisopropoxide, titanium n-butoxide, 2-ethylhexanoate Titanium, titanium di(acetylacetonate) oxide

Yttrium 2-ethylhexanoate [Y(OOCCH(C ₂ H ₅ )C ₄ H ₉ ) ₃ ], Yttrium nitrate, Yttrium isopropoxide, Yttrium cyclohexanecarboxylate (Y-napthenoate)

Sr strontium nitrate [Sr(NO ₃ ) ₂ ], strontium 2-ethylhexanoate, Sr(TMHD)

Co Cobalt cyclohexanecarboxylate, cobalt carbonyl, cobalt nitrate

Au triethylphosphine chloride alloy (I), triphenylphosphine chloride alloy (I)

B trimethyl borate, B-trimethoxyboroxine

K Potassium ethoxide, potassium tert-butoxide, potassium 2,2,6,6,-tetramethylheptane-3,5-dioate

Sodium 2,2,6,6-tetramethylheptane-3,5-dioate, sodium ethoxide, sodium tert-butoxide

Li 2,2,6,6-tetramethylheptane-3,5-dioate lithium, lithium ethoxide, lithium tert-butoxide

Cu Cu(2-ethylhexanoate)2, copper nitrate, copper acetylacetonate

Pd palladium nitrate (in aqueous ammonium hydroxide) (NH ₄ ) ₂ Pd(NO ₂ ) ₂ , palladium acetylacetonate, ammonium hexochloropalladium

Ir H ₂ IrCl ₆ (in 50% aqueous ethanol), iridium acetylacetonate, iridium carbonyl

Ag Silver nitrate (in water), silver nitrate, silver fluoroacetate, silver acetate, silver cyclohexanebutyrate, silver 2-ethylhexanoate

Cd Cadmium nitrate (in water), Cadmium 2-ethylhexanoate

Nb Niobium 2-ethylhexanoate

Mo (NH ₄ ) ₆ Mo ₇ O ₂₄ , Mo(CO) ₆ , molybdenum di(acetylacetonate) dioxide

Fe Fe(NO ₃ ) ₃ 9H ₂ O, iron acetylacetonate

Sn SnCl ₂ 2H ₂ O, tin 2-ethylhexanoate, Sn-tetra-n-butyltin, tetramethyltin

In In(NO ₃ ) ₃ xH ₂ O, indium acetylacetonate

Bi bismuth nitrate, bismuth 2-ethylhexanoate

Ru Ruthenium acetylacetonate

Zn Zinc 2-ethylhexanoate, zinc nitrate, zinc acetate

W Tungsten hexacarbonyl, tungsten hexafluoride, tungstic acid

In most cases where mixtures of metal precursors and/or metalloid precursors are deposited, the deposits are generally stoichiometric in relation to the relative proportions of metal and/or metalloid provided by the precursors in the reaction mixture. However, this relationship is neither precise nor completely predictable. Even so, it does not pose any significant problems in obtaining a coating or powder of the desired composition, because the relative amounts of chemical precursors required to obtain a coating or powder of the desired composition can be readily determined without the need for further analysis. Any one set of coating parameters is too much experimentation. Once the proportions of chemical precursors to obtain a coating or powder of the desired composition are determined under a set of coating parameters, the coating can be reproduced with highly predictable results. Thus, if it is desired that the coating or powder contain a particular predetermined ratio of the two metals, one can start with two chemical precursors containing the two metals in the predetermined stoichiometric ratio. If it is determined that the two metals will not deposit in the predetermined ratio, the relative amounts of the two precursor chemicals are adjusted until the desired ratio of metals is obtained in the deposited material. This empirically determined quantity will then be the basis for future deposition.

CCVD has the advantage of being able to deposit very thin, uniform layers that can be used as dielectric layers for buried capacitors and resistors. For buried resistors, the thickness of the deposited layer is usually at least about 40 Angstroms. The material can be deposited to any desired thickness; however, for CCVD to form resistive layers, thicknesses rarely exceed 50,000 Angstroms (5 microns). Typical film thicknesses are in the range of 100-10,000 Angstroms, most commonly in the range of 300-5000 Angstroms. The ability to deposit very thin films is an advantage of the CCVD method because the thinner the resistive layer, the higher the resistance and the less material (eg when using platinum). The thin coating also facilitates rapid etching during processing by which discrete resistors are formed.

Examples of coatings produced by CCVD include silica coatings obtained from a solution of tetraethoxysilane [Si(OC ₂ H ₅ ) ₄ ] in isopropanol and propane; Platinum coatings prepared from solutions of (CH ₃ COCHCOCH ₃ ) ₂ ] in toluene and methanol; nickel-doped LaCrO ₃ coatings prepared from solutions of lanthanum nitrate in ethanol, chromium nitrate in ethanol, and nickel nitrate in ethanol layer.

The resistance value of the resistor depends on the resistivity of the material and the length and cross-sectional area of the resistor. While very thin films are desirable from a material efficiency standpoint, thicker films may be required when the power load (current) is high. For higher power load requirements requiring thicker films, higher resistivity of the material is required, for example by using more doped metals as resistive materials.

New resistive materials can be deposited by CCVD and CACCVD, so CCVD and CACCVD processes can be used in combination with conventional or modified printed circuit board technology to form multiple very small discrete resistors. New types of resistive materials are formed by co-depositing conductive materials, especially metals such as platinum and nickel, with high resistive (dielectric) materials such as silicon dioxide by CCVD and CACCVD. It has been found that very small amounts of highly resistive materials (eg, about 0.1%-20% by weight) can greatly reduce the conductive properties of conductive materials. For example, platinum, although an excellent conductor, can be used as a resistor when co-deposited with 0.1 to about 5 wt. related. While applicants do not wish to be bound by theory, it is believed that when a conductor and a small amount of nonconductor are co-deposited by CCVD or CACCVD, the nonconductor typically deposits uniformly throughout the conductor as individual molecules or molecular nanoparticles.

For a resistive material (which is a mixture of a conductive metal and a small amount of dielectric material) to be deposited by CCVD or CACCVD, the metal must be capable of being deposited as a zero-valent metal in an oxygen-containing system. The criterion for the flame to bring the deposit to the zero-valent state is that the oxidation potential of the metal must be lower than the lower of the oxidation potentials of carbon dioxide and water at the deposition temperature. (At room temperature, water has a lower oxidation potential; at other temperatures, carbon dioxide has a lower oxidation potential.) Zero-valent metals that can be readily deposited by CCVD are those with oxidation potentials equal to or lower than silver. Thus, Ag, Au, Pt and Ir can be deposited by direct CCVD. Zero-valent metals with slightly higher oxidation potentials can be deposited using CACCVD processes that provide a more reducing atmosphere. Ni, Cu, In, Pd, Sn, Fe, Mo, Co and Pb are best deposited by CACCVD. Metals here also include alloys, which are mixtures of these zero-valent metals. Preferred dielectric materials that can be co-deposited with zero-valent metals are metal oxides or metalloid oxides such as silica, alumina, chromia, titania, ceria, zinc oxide, zirconia, phosphorus oxide ( phosphorous oxide), bismuth oxide, commonly used oxides of rare earth metals, and mixtures thereof. Silicon, aluminum, chromium, titanium, cerium, zinc, zirconium, magnesium, bismuth, rare earth metals, and phosphorus all have such high oxidation potentials that if any of the above Metals, which will deposit in their zero-valent state, and dopants will deposit as oxides. Therefore, even without the use of a flame, the dielectric material needs to have a high oxidation, phosphating, carbonizing, nitrifying, or boriding potential to form the desired two phases.

Also, for more oxygen reactive metals and metal alloys, CACCVD is the preferred process. Even though the metal can be deposited as a zero valent metal by direct CCVD, it is preferable to provide a controlled atmosphere (ie CACCVD) if the substrate to be deposited thereon is susceptible to oxidation. For example, copper and nickel substrates are easily oxidized and are preferably deposited on these substrates by CACCVD.

Another resistive material that can be deposited in a thin layer on a substrate by CCVD is a "conducting oxide". Especially Bi ₂ Ru ₂ O ₇ and SrRuO ₃ are conductive oxides that can be deposited by CCVD. Although these materials are "conducting," their electrical conductivity is very low when deposited in an amorphous state; thus, a thin layer of these mixed oxides can be used to form discrete resistors. Like conducting metals, "conducting oxides" can be doped with dielectric materials (such as oxides of metals or metalloids) to increase their resistivity. These mixed oxides can be deposited as amorphous or crystalline layers, with amorphous layers tending to be deposited at low deposition temperatures and crystalline layers tending to be deposited at higher deposition temperatures. For use as resistors, generally preferred are amorphous layers, which have a higher resistivity than crystalline layers. Thus, while these materials are classified as "conducting oxides" in their normal crystalline state, amorphous oxides (even in undoped form) can yield good resistance values. In some cases, it may be necessary to form a low-resistance (1-100Ω) resistor, and a conductivity-enhancing dopant, such as Pt, Au, Ag, Cu or F, can be added. These homogeneously mixed dielectric materials, if doped with a dielectric material (such as an oxide of a metal or metalloid) to increase the resistivity of the conductive oxide, or to reduce the resistivity of the conductive oxide with a conductivity-enhancing material, Or the conductivity-enhancing material is usually present in an amount of about 0.1-20% by weight of the resistive material, preferably at least about 0.5% by weight.

There are many other "conductive materials" which, while conducting electricity, have sufficient resistivity to form the resistors of the present invention. Examples thereof include yttrium barium copper oxide and La _1-x Sr _x CoO ₃ , 0≤x≤1, eg x=0.5. In general, any mixed oxide that has superconducting properties below a critical temperature can be used as a resistive material above that critical temperature. Such a variety of resistive materials may be deposited by suitable selection of precursors from those described herein above.

To obtain a metal/oxide resistive material film, a precursor solution is provided that contains both a metal precursor and a precursor of an oxide of the metal or metalloid. For example, to obtain a platinum/silica film, the deposition solution contains a platinum precursor such as platinum(II) acetylacetonate or platinum(II) diphenyl-(1,5-cyclooctadiene)[Pt(COD )] and silicon-containing precursors (such as tetraethoxysilane). These precursors are mixed roughly in proportion to the metal to be deposited and the metal or metalloid (which will form an oxide); however, the exact ratio of metal to oxide must be determined empirically for any desired ratio. Proportion. Therefore, the precursor solution for forming a resistive film of the present invention contains a metal-forming precursor and an oxide-forming precursor in a weight ratio of about 100:0.2 to 100:20.

Similarly, when depositing a conductive oxide to form a layer of resistive material, the precursors for each metal (Bi and Ru, and Sr and Ru) must be provided in the proper ratios to obtain the correct stoichiometric amount of the conductive oxide . Likewise, some experimentation may be done to determine the precise proportions of precursors to be supplied at any particular deposition condition to obtain the desired stoichiometric amount of mixed oxide. In addition, when the conductive oxide is doped with a dielectric metal oxide or metalloid oxide to increase the resistivity of the material to be deposited, or with a conductivity-enhancing material to lower the resistivity of the material to be deposited, Additional precursors may be added to obtain small amounts of metal oxide or metalloid oxide, such as 0.1-20% by weight of the deposited doped conductive metal oxide, preferably at least about 0.5% by weight. metal oxides or metalloid oxides.

Either of the two platinum precursors mentioned above is soluble in toluene. Dissolution of the platinum precursor can be facilitated by sonification. Tetraethoxysilane dissolved in methanol, isopropanol or toluene can be conveniently added to the solution of the platinum precursor to form the precursor solution. This precursor solution can then be further diluted to the desired concentration with propane or other organic solvent.

Generally, for transport, storage and processing, the precursor chemicals are dissolved in common liquid organic solvents such as toluene, isopropanol, methanol, xylene, and mixtures thereof, (all precursor chemicals ) in a concentration of about 0.25-20% by weight, preferably at least about 0.5% by weight, and usually up to about 5% by weight. In general, for shipping and processing, it is desirable to provide the concentrate in concentrated form to minimize cost and flammable liquid content. At the same time, stability must be considered, especially low-temperature stability (eg, down to -20° C.), in order to avoid precipitation of over-concentrated solutions. During deposition, the precursor solution is typically further diluted, for example in propane, to a concentration (of all precursor chemicals) of about 0.005-1.0% by weight, preferably about 0.05-1.0% by weight ), more preferably not more than about 0.6% by weight.

One of the most important metals that can be deposited by CACCVD in doped or undoped form is nickel. Nickel is inexpensive and can be etched selectively relative to conductive metals such as copper. An important precursor for depositing zero-valent nickel by CACCVD is nickel nitrate. Nickel can be deposited from an ammonia solution of nickel nitrate. However, as mentioned above, it is preferred to deposit from a liquid which is in a near supercritical state. Advantageous nickel nitrate supports for this purpose include liquefied ammonia or liquefied nitrous oxide ( _N2O ). Nitrous oxide can be pressurized to 700-800 psi for liquefaction. Ammonia can be liquefied by pressurization and/or low temperature. It has been found that whether the carrier is liquefied ammonia or liquefied nitrous oxide, it is preferred to add a small amount of water, i.e. up to about 40% by weight, preferably about 2-20% by weight of water, (balance It is liquefied ammonia or liquefied nitrous oxide, about 60-100% (weight)). Water can raise the supercritical point of liquefied ammonia or liquefied nitrous oxide. This makes it easier to operate well below the supercritical point so that changes in viscosity and density are not encountered. Furthermore, the addition of water reduces the instability of the solution. (However, it should be understood that in some cases the deposition can be carried out in liquefied ammonia or liquefied nitrous oxide without the addition of water.) In such a nickel deposition solution, the nickel precursor is combined with the nickel dopant Any precursors of are generally present in small amounts, ie, about 0.001-2.5% by weight. Presently preferred dopants for nickel are nickel phosphorous and/or nickel phosphorous oxides, such as nickel phosphate. It is believed that when using phosphorus-containing precursors such as phosphoric acid, the dominant dopant species is nickel phosphate. Precursor solutions of water and liquefied ammonia or _N2O as carrier co-solvents are advantageous because of the absence of carbon, which could lead to carbon deposition.

In preparing a nickel nitrate precursor solution in a liquefied ammonia carrier, it may be convenient to predissolve the precursors of nickel nitrate and any dopants in an ammonium hydroxide solution and then mix this solution with liquefied ammonia.

The resistive materials described herein can be fabricated as resistors, either as embedded resistors or as resistors on the surface of printed wiring boards in integrated circuits or other electronic applications. This is usually accomplished by forming a resist pattern on the layer of resistive material using a photoresist and using a suitable etchant to remove the resistive material in areas not covered by the resist pattern. For metal/oxide resistive material layers, the etchant of choice is an etchant for the metallic component of the resistive material. Typically these etchants are acids or Lewis acids such as _FeCl3 or _CuCl2 for copper. Nitric acid and other inorganic acids such as sulfuric, hydrochloric, and phosphoric acids can be used to etch nickel, various other metals that can be deposited, and conductive oxides.

Aqua regia can be used to etch precious metals such as platinum. Aqua regia is an extremely corrosive acid mixture that is used in this invention to etch metals, especially precious metals such as platinum and gold. Au can also be etched in a potassium iodide/iodine (KI/I ₂ ) solution. Because CCVD uses a flame, it tends to produce oxides, and only less reactive metals (ie, metals with a low oxidation potential) tend to deposit as metals rather than oxides. The easiest to deposit are precious metals such as platinum and gold. Of course these metals are expensive, but the advantage of CCVD is that it can be used to deposit very thin but uniform films. Therefore, the use of CCVD to deposit thin layers of noble metals is feasible in many cases. Moreover, since noble metals are non-oxidizing, their economic benefits for high-quality electronic applications can be easily demonstrated.

Furthermore, although noble metals are conductors, it has been found that when noble metals are deposited with relatively small amounts of oxides such as silicon dioxide or aluminum oxide, the deposited materials are highly resistive. Therefore, metals (such as platinum) containing small amounts (such as 0.1%-5%) of oxides can be used as resistors in printed wiring boards. These materials can be deposited in layers on printed circuit boards and then processed using printed circuit board technology to obtain discrete resistors.

However, noble metals are difficult to etch due to their non-reactive nature, which is required in many processes for making printed wiring boards. Aqua regia is used as an etchant for metals (especially precious metals) in printed circuit board technology.

Aqua regia is prepared from two known acids: 3 parts concentrated hydrochloric acid (12M HCl) and 1 part concentrated nitric acid ( _16MHNO3 ). Thus, the molar ratio of hydrochloric acid to nitric acid is 9:4, although slight variations in this ratio (ie, 6:4 to 12:4) are acceptable for the etching purposes of the present invention. Because aqua regia is corrosive and has a limited lifespan, it is not commercially available and must be prepared on site. To make it less corrosive, aqua regia can be diluted with water up to about a 3:1 ratio of water to aqua regia. Of course dilution with water will increase the etch time, but good etch times for platinum can be achieved with 33% aqua regia solution. Of course, more reactive metals such as copper can be etched more easily with aqua regia. On the other hand, noble metals such as platinum are not etched by many materials suitable for etching copper such as _FeCl3 or _CuCl2 , so a variety of selective etches can be used in forming printed wiring boards.

The rate of etching depends on several factors including the strength and temperature of the aqua regia. Aqua regia is preferably freshly prepared. Generally, aqua regia etching is done in the 55-60° range, although this will vary with the application.

The following discussion of discrete resistor formation assumes the use of platinum-based resistor materials, since platinum/silicon dioxide is currently the preferred CCVD-deposited resistor material. It should be understood, however, that other resistive materials may be substituted, including films of the metal/oxides and conductive oxides described above. Also, in the techniques for selectively etching copper and platinum-based resistive layers described below, it should be understood that there are a variety of selective etchants that can be used with the various conductor/resistive material combinations of the present invention.

In its simplest form, the resistor 400 of the present invention is simply a piece or strip of thin layer resistive material 401 (such as 4c and 4d) on an insulating substrate 402, which is arranged with a copper strip 403 for contact at each end. , to provide an electrical connection between the resistor and the electronic circuit. Substrate 402 may be a flexible sheet (such as a polyimide sheet), a rigid epoxy/fiberglass sheet, or even a liquid crystal sheet. A suitable substrate for many applications is a film of an organic polymer such as polyimide having a thickness of about 10 microns or less. It was found herein that, after optimizing the deposition parameters, CCVD can apply a layer of resistive material to an insulating substrate such as polyimide without burning or deforming the substrate. Depositing the resistive material layer directly on the insulating substrate generally results in good adhesion of the resistive material layer to the insulating substrate. In general, this bonding is superior to prior art techniques that use adhesives to bond resistive materials to substrates. The discrete resistor 400 is formed as follows: a thin layer of resistive material 401 (4a) is deposited on an insulating substrate 402 by CCVD to form the structure of FIG. 4a. A chemically resistant photoresist, such as Laminar 5038 aqua regia resistant resist (in the case of platinum etching) sold by Morton Electronic Materials, is applied to the exposed surface of the resistive material and patterned using conventional photoimaging techniques . In general, a resist that can withstand very highly acidic conditions such as gold plating conditions is suitable for use as a resist when etching is performed with aqua regia. The exposed portion of the resistive material layer is thus etched away, (etching using aqua regia in the case of a noble metal based resistive material) leaving blocks or strips of resistive material 401 (4b) to form the structure of FIG. 4b. Then, connecting copper sheets 403 may be applied to both ends of the resistive material strip 401 to form the resistor 400 in FIG. 4c.

Photo-imaging techniques are then used to form blocks 401 of thin layer resistive material and blocks 403 of electrically connecting conductors, preferably referring to Figures 5a-5c. Figure 5a shows a three-layer structure 409 comprising an insulating substrate 402, a resistive material layer 401 (5a) (such as Pt/silicon dioxide) formed by CCVD according to the present invention, and a resistive material layer 401 (5a) formed by CCVD or other techniques (such as electrolytic Electroplating) formed conductive layer 403 (5a) (such as copper).

The structure 409 of Figure 5a can be patterned using photoimaging techniques in one of two two-step processes. In one approach (see Figure 5b), the layer of conductive material 403 (5a) is covered with a resist, the resist is patterned using photoimaging techniques, and the exposed areas of the resist are then treated with, for example, aqua regia. Etching away the layer of conductive material and the underlying layer of resistive material results in the structure of Figure 5b having a patterned block of resistive material (401(5b)) and a patterned block of conductive material (403(5b)). Next, a second resist is applied, photoimaged and developed. At this time, only the exposed portion of the conductive material block (403(5b)) is etched away by the etchant, thereby obtaining the resistive structure 400 of FIG. 4c. The etchant is capable of selectively etching the conductive layer without etching the block of resistive material, for example _FeCl3 or _CuCl2 in the case of Cu as the conductive material layer and Pt/silicon dioxide as the resistive material. In another method (referring to Fig. 5 c), form the resist layer with pattern, use for example FeCl ₃ etch to remove the exposed part of conductive material layer 403 (5a), form a resist layer with pattern again, then use Aqua regia etching removes the exposed area of the resistive material layer (401(5b)) to form an electrical contact 403 and obtain the resistive structure 400 of FIG. 4c. Using either method, discrete sheet resistors 400 can be formed using conventional photoimaging techniques commonly used to form printed wiring.

Another way to form discrete resistors is to start with a two-layer structure with a layer of resistive material (such as Pt/silicon monoxide) on an insulating substrate as shown in Figure 4a, using a photoresist method, on the substrate A block or a strip of discrete resistor material is formed on it to obtain the structure shown in Figure 4b. Next, a layer of conductive material (such as copper) is formed on the resistance blocks or strips, for example by electrolytic plating, to obtain the structure shown in FIG. 5b. A further layer of photoresist is applied and imaged, and the exposed portions of the conductive material are then etched away to leave conductive electrical connection blocks 403 and result in a resistive structure 400 as shown in Figure 4c.

Although the resistor 400 of FIG. 4c can be located on the surface of the printed wiring board device, in most cases these resistors are buried in a multilayer printed wiring board, as shown in FIG. 6, where an insulating substrate 402 (such as polyamide Resistor 400 formed on imine) is buried in an additional layer of insulating material 420 such as epoxy/fiberglass prepreg.

Figures 9a-g show cross-sectional views of circuitization methods, which start with a conductive foil 900 (such as copper foil) and deposit a layer of resistive material 905 by CCVD or CACCVD. This two-layer structure is represented by Figure 9a. The thickness of the copper foil used in this method is usually about 3-50 microns.

Then, photoresist layers 910 and 915 are applied to both sides of the bilayer structure. The photoresist 910 covering the resistive material layer 905 is exposed to patterned actinic radiation, while the photoresist 915 covering the conductive foil 900 is blanket-exposed to actinic radiation. The photoresist is then developed to obtain the structure shown in FIG. 9 b . The resistive material layer 905 is covered with a patterned photoresist layer, and the conductive foil is protected by a blanket-exposed photoresist layer 915 .

As shown in FIG. 9c, the resistive material layer 905 is then selectively etched away from the areas where the photoresist 910 has been removed. The remaining photoresists 910 and 915 are then removed.

Thereafter, as shown in Figure 9d, an organic stack 920 is applied to the resistive material side of the structure. The laminate protects the now patterned layer of resistive material 905 from further processing and acts to support the pieces of resistive material layer 905 when portions of the conductive foil are subsequently removed from the other side of the resistive material layer.

A photoresist layer 925 is then applied to the conductive foil 990 . This was imaged and developed with patterned actinic radiation to give the structure shown in Figure 9e. Thereafter, the conductive foil 900 is etched with an etchant that selectively etches the conductive foil 900 but not the resistive material layer 905, leaving the structure shown in Figure 9f. The photoresist 925 is removed, leaving the resistive structure shown in Figure 9g. This structure can then be embedded in a dielectric material (not shown in the figure).

As a variation on this method, it should be noted that if the etchant used is capable of selectively etching the resistive material layer 905 without etching or only partially etching the conductive foil 900, then it is not necessary to apply the etch layer 915 (FIGS. 9b and 9c). .

When "etching" is mentioned herein, the term is used not only in the ordinary sense in the art (in which a strong chemical dissolves a layer of material, such as nitric acid dissolves nickel), but also to include physical removal (such as laser removal and removal through insufficient adhesion). In this regard, and in accordance with one aspect of the invention, it is contemplated that resistive materials deposited by CCVD or CACCVD, such as doped nickel and doped platinum, are porous. Its porosity is believed to allow liquid etchant to diffuse through the layer of resistive material, physically breaking the adhesion between the layer of resistive material and the underlying layer.

For example, referring to Figures 9b and 9c, if the conductive foil layer 900 is copper and the resistive material layer 905 is doped platinum (eg, Pt/silicon dioxide) or doped nickel (eg, Ni/PO ₄ ), chloride can be used. copper to remove the exposed portion of the resistive material layer. Cupric chloride does not dissolve Pt or Ni, but the porosity of the resistive material layer allows copper chloride to reach the copper beneath it. A small portion of copper is dissolved, and the exposed portion of the resistive layer 905 is physically ablated. This physical exfoliation occurs before the copper chloride significantly etches the underlying copper layer 900 .

For the same reason, the porosity of resistive materials deposited according to the invention can be removed by exfoliation. For example, a platinum layer on a polyimide substrate can be etched away with an etchant such as that described above for removing a resistive layer from a conductive copper substrate, especially mineral acids such as hydrochloric acid, sulfuric acid, and acidic Copper chloride (acidic cupric chloride). Therefore, in methods such as those described so far using ordinary photoresist technology, discrete resistors can be formed by etching a thin film of resistive material on an insulating substrate such as a polyimide film.

If the layer of conductive material 900 is copper, it is sometimes advantageous to use copper foil that has been oxidized; oxidized copper foil is commercially available. The advantage of oxidized copper foil is that dilute HCl solution (eg 1/2%) dissolves copper oxide but not zero valent copper. Thus, HCl can be used for exfoliation etching if the layer of resistive material is porous such that dilute HCl solution diffuses through it. Dissolving the surface copper oxide will break the adhesion between the copper foil and the layer of resistive material. As noted above with respect to the method shown in Figures 9a-9g, the use of an etchant that does not attack the foil eliminates the need for a protective photoresist layer 915 (Figures 9b and 9c).

To minimize processing steps, the applied photoresist can be embedded in the material, such as Morton International's permanent resist. If the etchant does not etch the conductor or only partially etches the conductor, both sides of the material can be processed simultaneously. Specifically, only the photoresist on the resistive material side needs to be able to be embedded, while the photoresist on the conductor side can be removed as a final processing step. Alternatively, the photoresist used on the conductive material side can be selected so that it is not removed by the special stripper used to remove the photoresist on the resistive material side. A photoresist that can be embedded reduces the loss of tolerance caused by underetching of the resistive material, since the underetched material flakes off after the photoresist is removed.

It can be shown that when layers of porous resistive material such as doped platinum and doped silicon dioxide are used with certain etchants, the etching process is a physical exfoliation process. This is due to flakes of resistive material found in the etchant bath. Because of this, exfoliated resistive material can be separated from the etchant bath by physical means (eg, filtration, settling, centrifugation, etc.). This is especially convenient for recovering expensive materials such as platinum.

To be practically removable by the exfoliation technique, the layer of resistive material must generally be sufficiently porous to an etchant that does not dissolve the resistive material but is sufficient to corrode the underlying material surface so that the interfacial adhesion is lost and peeled off within about 2 to 5 minutes conductive material. At the same time, the etchant must be such that it does not significantly corrode the underlying material (eg, copper foil) during etching, since this corrosion may result in excessive underetching or loss of mechanical strength (ie, reduced processability).

Thus, Fig. 10a shows a structure relating to the above, a conductive layer 1000, such as copper; an intermediate etchable layer 1002, such as copper oxide; a porous layer 1004 of resistive material through which etchant can penetrate and Dissolves the intermediate layer without significantly eroding the conductive layer. Referring to FIG. 10b, a patterned resistive layer 1006 is formed by exposure and development; then, referring to FIG. 10c, the resistive layer 1004 is etched and peeled off to form a patterned resistive layer. This peeling is performed as follows: the resistive layer 1004 is contacted with an etchant, The etchant penetrates through the porous resistive layer and corrodes the intermediate layer 1002 so that the overlying resistive layer is mechanically stripped away.

Although copper oxide is a suitable interlayer 1002 from the standpoint of being able to selectively etch the underlying copper conductive layer 1000 , copper oxide is not a preferred material for the interlayer 1002 . It has been found that when a resistive material such as silicon dioxide doped platinum is deposited directly on copper or copper oxide, the copper and/or copper oxide has a tendency to react with the resistive material such that the resistivity of the resistive material May be unpredictable. Therefore, it is preferable to apply the intermediate layer 1002 to the conductive foil layer 1000 before applying the resistive material by CCVD or CACCVD, the material of the intermediate layer should be such that the conductive material in the foil layer 1000 does not diffuse into the resistive material layer 1004 .

The requirements for the material used for the intermediate layer 1002 must be such that the etchant can sufficiently degrade the intermediate layer to peel off the resistive material layer 1004 . It is preferred that the etchant minimize or not degrade the conductive layer 1000 . For example, there may be chemicals that etch the intermediate layer but do not react with the conductive layer 1000 . However, even if a chemical substance can etch the material forming the intermediate layer 1002 and can also etch the material forming the conductive layer 1000, it is generally still possible to use the etchant by controlling the etching conditions (including time) to etch the intermediate layer 1002 without Conductive layer 1000 will be significantly etched away. For example, if the conductive layer 1000 is copper and the intermediate layer 1002 is nickel, copper chloride, which will etch away both nickel and copper, is a suitable etch resist, as long as the etch conditions are controlled so that the very thin nickel layer is significantly etched away, without significantly etching away much thicker copper layers. In addition, the material of the intermediate layer 1002 must maintain good electrical contact between the conductive layer 1000 and the resistive material layer.

One choice of material for the intermediate layer 1002 is a metal, such as nickel, which prevents interaction between the copper and the layer of resistive material 1004 by providing a barrier layer between conductive layers (eg, copper). Nickel can be deposited on copper, for example by electroplating. Typically, the nickel interlayer is about 2-6 microns, although the thickness is not believed to be particularly critical.

Another choice for the material of the intermediate layer 1002 is ceramics, such as silicon dioxide or oxides of other metals or metalloids. Such an intermediate layer may be deposited by CCVD as described above prior to depositing the resistive material layer 1004 . Although most ceramic materials such as silicon dioxide are electrically insulating (dielectric), dielectric materials can still be used as intermediates if deposited in sufficiently thin layers (e.g., about 15-50 nanometers in average thickness). The barrier layer 1002 does not significantly disrupt the electrical contact between the conductive layer 1000 and the resistive layer 1004 . (When discussing interlayer thickness, the discussion is about the average thickness, since the thickness usually varies from place to place, depending on factors such as substrate roughness and deposition conditions.) The net effect is an etchable leaky interlayer, which is effectively composed of buffer layers.

Suitable etchants for silicon dioxide include ammonium bifluoride, fluoroboric acid, and mixtures thereof, if silicon dioxide is used as the interlayer. A particularly suitable etchant for silicon dioxide, if silicon dioxide is used as the interlayer, is an aqueous solution containing 1.7% by weight of ammonium bifluoride and 1.05% by weight of fluoroboric acid. Other materials may be added to the mixture of these two components.

In the case of silicon dioxide, sufficiently nanoporous or defective coatings are required to enable nanoscale direct contact points between resistors and conductors. The size of these contacts can be 1-100 nanometers, forming an area of 0.05%-10%, thereby allowing the resistive feature size to be further reduced to micron-scale resolution while still providing a good electrical connection. This still sufficiently reduces material interaction. Alternatively poor insulators, semiconducting or conducting composite ceramic or polymer materials can be used, in which case these can be thicker. Also, in this regard, the rougher the substrate surface, the thicker the interlayer, possibly because rougher substrate surfaces tend to result in more porous interlayer coatings. That is, we believe that the rougher the surface of the substrate, the greater the number of pinholes produced in the intermediate coating through which electrical contact can be maintained.

Other oxides that can be used as interlayers include zinc oxide, strontium oxide, and tungsten oxide. Each of these oxides can be deposited by CCVD using the zinc, strontium and tungsten precursors described above. Each of these oxides can be applied to copper substrates by CCVD at temperatures low enough that the copper does not oxidize. Each of these oxides can be applied at relatively low cost.

Zinc oxide is a particularly promising interlayer material because it is an electrical semiconductor. Therefore, it is able to provide good electrical continuity between the conductive metal (such as copper) and the resistor. Zinc oxide (and other oxides) can be doped to increase electrical conductivity. In addition, zinc oxide can be etched with hydrochloric acid.

Strontium oxide and tungsten oxide are etchable with strong bases such as KOH.

Detailed ways

Now the present invention will be described in more detail by specific examples.

Example 1

A layer of Pt/ _SiO2 resistance material is deposited on polyimide by CCVD, and the deposition conditions used are as follows:

Solution preparation: 1.23 g of Pt(COD)

                                                                       

         0.43 g TEOS (1.5% by weight Si in toluene)

150 grams of propane

Deposition conditions: Solution flow rate: 3 ml/min

Deposition time: about 18 minutes for a 5″×6″ substrate

Deposition Times (#of passes): 6

Deposition temperature: 500°C

         Autotransformer: 3.0A

Oxygen flow rate at the tip: about 2900 ml/min

The resistance value of the sample obtained from the above deposition conditions was about 17 ohms per square.

This is an example of a 65% concentrated solution containing 2.5% by weight _SiO2 . Variables that can be varied include the amount of Pt(COD) and TEOS added in proportion to bring the solution concentration to 100% (e.g., 1.89 grams of Pt(COD) and 0.65 grams of TEOS (1.5% by weight Si)), and the amount that can be added to vary The amount of SiO obtained _{is 2} % by weight of TEOS (usually 0.5-5% by weight for this purpose).

Example 2

In some cases, it is desirable to deposit certain materials on oxidation-sensitive substrates without oxidizing the substrate. This can be done using CACCVD techniques, an example being the deposition of the dielectric compound _SrTiO3 on Ni. This deposition uses a conventional CCVD nozzle placed in a thimble that provides an inert or reducing gas around the flame. As shown in Figure 8, a jacketed nozzle was then placed inside the quartz tube to prevent air from reaching the substrate during deposition. For this CACCVD flame, the flammable solution flows along the needle, the oxygen flows along the tip, and the hydrogen flows along the squib as in the CCVD process. A high flow of inert gas (such as argon or nitrogen) or reducing gas (such as 90-99.5% argon/10-0.5% hydrogen) flows around the flame along the sleeve. For very small samples, a sidearm purged with an inert or reducing gas is part of the quartz tube so that the heated sample is cooled in a controlled atmosphere after deposition, preventing oxidation at that point. According to EDX and XRD analysis, this method enables the deposition of SrTiO ₃ on Ni without the formation of NiO or the deposition of carbon. Early experiments showed that using a solvent with a low carbon deposition potential, such as methanol, was superior to using toluene. Carbon is deposited on the substrate when toluene is used. The ideal process parameters so far are as follows:

Preparation of solution: 0.82 g strontium 2-ethylhexanoate (1.5% by weight Sr in toluene)

      0.73 g Ti-diisoacac (Ti-di-i-acac)

  (Containing 0.94% (weight) Ti in toluene)

                                               

100 grams of propane

Deposition conditions: solution flow rate: 2 ml/min

Deposition time: 15 minutes (varies within 10-15 minutes)

Deposition temperature: ~950°C (variable within 800-1050°C)

Autotransformer: 1.9A (varies between 1.9-2.25A)

    Pyrogen hydrogen flow rate: ～1926 ml/min

(down to 550ml/min)

    Oxygen flow rate at the tip: ～1300ml/min

(Varies between 500-2322ml/min)

     Reducing gas mixture: 0.5-10% hydrogen/balance argon

    Reducing gas flow rate: 58 liters/min

Example 3

Deposited on a 200TAB-E polyimide substrate using a solution of 0.760 g Ni(NO ₃ )H ₂ O and 0.30 g H ₃ PO ₄ in 400 mL 6M NH ₄ OH using the equipment shown in Figure 7 Phosphate-doped nickel films. The solution was passed through a 22-gauge (ga.) stainless steel needle with a 22 micron ID (0.006″ OD) fused silica capillary insert (3 mm long) at the tip 722 at a flow rate of 0.50 sccm. Hydrogen was injected at 1.20 lpm. Velocity flows through surrounding tube 738. Hydrogen flows through tube 744 surrounding 738 at a rate of 756 sccm. Oxygen flows through tube 752 surrounding 744 at a rate of 1.40 lpm. Argon flows through outer tube 768 at a rate of 28.1 lpm. All The gas flow starts before the flame is manually ignited. Generally, the argon flow has to be reduced in order for the flame peak to ignite the internal nozzle. Then return the argon flow to its original setting. Once lit, the long flame is no longer required Or ignite the source to keep the combustion. The temperature of the gas above the deposition point is about 1 mm is 500 ° C. At a distance of 2 mm from the nozzle guard ring, the substrate is scanned horizontally at a rate of 20 "/min and a step of 0.0625 " The upper 3.5" x 3.5" area is scanned once across. This scan movement takes a total of 12 minutes.

The deposited phosphate-doped nickel layer had a linear resistance of 115 ohms/inch.

This deposition was repeated using a solution without phosphoric acid for comparison. The resistance value of the nickel layer was 5 ohms/inch.

Example 4

Bi ₂ Ru ₂ O ₇ was deposited using the following chemistries and process parameters:

Precursor solution:

0.0254% by weight Bi in 2-ethylhexanoate + 0.0086% by weight Ru in acetylacetonate + 1.8026% by weight methanol + 15.0724% by weight toluene + 83.0910% by weight propane .

parameter:

Flow rate of precursor solution: 3 ml/min.

Nipple oxygen flow: 4 liters/minute.

Autotransformer: 2.30A.

No rear cooling.

Deposition temperature: 250-650°C.

Coating amorphous Bi ₂ Ru ₂ O ₇ at a gas temperature of 400° C. has a resistance value below 7200 microohm·cm (μΩ·cm); this is the best way to date. Propane and toluene were used as solvents. To prepare concentrated or dilute solutions for deposition, toluene in the range of 1-35% by weight can be used. Propane in the range of 99-65% by weight can also be used. By changing the solvent weight percentage, the concentration of solute (bismuth 2-ethylhexanoate and ruthenium acetylacetonate) can be adjusted accordingly. The flow rate of the precursor solution is in the range of 1-5 ml/min.

Example 5

_SrRuO3 was deposited using the following chemistry and process parameters:

Precursor solution:

0.0078 wt% Sr in 2-ethylhexanoate + 0.0090 wt% Ru in acetylacetonate + 12.7920 wt% toluene + 87.1912 wt% propane.

parameter:

Flow rate of precursor solution: 3 ml/min.

Nipple oxygen flow: 4 liters/minute.

Autotransformer: 2.75A.

No rear cooling.

Deposition temperature: 300-650°C.

Coating amorphous SrRuO ₃ at a gas temperature of 400° C. has a resistance value below 5400 microohm·cm (μΩ·cm); this is by far the best way. Propane and toluene were used as solvents. To prepare concentrated or dilute solutions for deposition, toluene in the range of 1-35% by weight can be used. Propane in the range of 99-65% by weight can also be used. By varying the solvent weight percentages, the concentrations of the solutes (strontium 2-ethylhexanoate and ruthenium acetylacetonate) can be adjusted accordingly. The flow rate of the precursor solution is in the range of 1-5 ml/min.

Example 6

Method One: Forming a Single Discrete Resistor

According to the method (of Example 1), a 200 nm thick platinum/silicon dioxide layer (Pt:SiO ₂ , 97.5:2.5) was deposited on a 25 μm thick polyimide sheet. A layer of photoresist, Laminar 5000 series resist sold by Morton International Electronics Materials, was laminated on the platinum layer. The resist layer was covered with a photo tool, and the uncovered portion of the resist layer was exposed to 70 millijoules of UV light. The unexposed resist was then removed by developing with a 1% sodium carbonate monohydrate solution at 80 F at about 25 psi using a conveyorized spray developer, adjusting the dwell time so that the breakpoint occurred at the length of the chamber ( chamber length) 40%-50%, followed by multiple spray rinses with tap water and deionized water.

Next, the sheet was exposed to 50% aqua regia solution (500 mL H ₂ O + 125 mL HNO ₃ + 375 mL HCl.) at 50°C for a sufficient time to remove all Pt/ _SiO2 material, thus forming discrete resistors.

Example 7

Method Two: Form a single discrete resistor with a copper connection circuit

According to the method (of Example 1), a 200 nm thick platinum/silicon dioxide layer (Pt:SiO ₂ , 97.5:2.5) was deposited on a 25 μm thick polyimide sheet. Copper was then plated directly onto the surface of the Pt/ _SiO2 layer to a thickness of 12 microns using a commercially available acidic copper plating bath, applying standard commercially available plating parameters. A layer of photoresist, a Laminar 5000 series resist sold by Morton International Electronics Materials, is laminated on the plated copper layer. The resist layer is covered with an optical tool, and the uncovered portion of the resist layer is exposed to 70 mJ of UV light. The unexposed resist was then removed by developing with a 1% sodium carbonate monohydrate solution at 80 F at about 25 psi using a conveyorized spray developer, adjusting the dwell time so that the breakpoint occurred 40%-50% of the length of the developing chamber % position, followed by multiple jet rinses with tap and deionized water.

Next, expose the sheet to 50% aqua regia solution (500 mL _H2O + 125 mL _HNO3 + 375 mL HCl.) at 50°C for a time sufficient to remove all plating in those areas where the resist has been removed Copper and Pt/SiO ₂ materials, thus forming electronic circuit patterns. Use a conveyorized spray resist stripper to remove the photoresist with a 3% NaOH solution at 130 F at about 25 psi, adjusting the dwell time so that the breakpoint occurs at 40%-50% of the length of the strip chamber, followed by Multiple jet rinses with tap and deionized water.

A layer of photoresist is laminated onto the circuitized electronic pattern, which is the Laminar 5000 series resist sold by Morton International Electronics Materials. The resist layer was covered with an optical tool, and the uncovered portions of the resist layer (all areas except the discrete resistor areas) were exposed to 70 mJ of UV light. Then use a conveyorized spray developer at about 25psi to develop with a 1% sodium carbonate monohydrate solution at 80°C to remove the unexposed resist, adjusting the dwell time so that the break point occurs at 40%-50 of the length of the developing chamber % position, followed by multiple jet rinses with tap and deionized water. The exposed copper areas were then etched in a commercial cupric chloride etchant to remove only the copper leaving the Pt/ _SiO2 exposed and uncorroded to form a resistor connected at each end with copper circuit traces. Use a conveyorized spray resist stripper to remove the photoresist with a 3% NaOH solution at 130 F at about 25 psi, adjusting the dwell time so that the breakpoint occurs at 40%-50% of the length of the strip chamber, followed by Multiple jet rinses with tap and deionized water.

Example 8

Method Three: Forming a Single Discrete Resistor with a Copper Connection Circuit

According to the method (of Example 1), a 200 nm thick platinum/silicon dioxide layer (Pt:SiO ₂ , 97.5:2.5) was deposited on a 25 μm thick polyimide sheet. Then, copper was plated directly onto the surface of the Pt/ _SiO2 layer to a thickness of 12 μm using a commercially available acidic copper plating bath and plating parameters. A layer of photoresist, a Laminar 5000 series resist sold by Morton International Electronics Materials, is laminated on the plated copper layer. The resist layer is covered with an optical tool, and the uncovered portion of the resist layer is exposed to 70 mJ of UV light. The unexposed resist was then removed by developing with a 1% sodium carbonate monohydrate solution at 80 F at about 25 psi using a conveyorized spray developer, adjusting the dwell time so that the breakpoint occurred 40%-50% of the length of the developing chamber % position, followed by multiple jet rinses with tap and deionized water.

The copper was then etched in a commercial etchant copper chloride, exposing the Pt/ _SiO2 . The resist was stripped and a new layer of photoresist (Laminar 5000 series) was applied using an industry standard vacuum lamination process. A second photomask having a line width 2 mils wider than the first pattern was used to expose the second pattern using the same exposure parameters as were used in the first resist exposure operation.

Next, the sheet was exposed to 50% aqueous regia solution (500 mL _H2O + 125 mL _HNO3 + 375 mL HCl) at 50 °C for a time sufficient to remove all exposed Pt/ SiO ₂ material, thus forming electronic circuit patterns. Use a conveyorized spray resist stripper to remove the photoresist with a 3% NaOH solution at 130 F at about 25 psi, adjusting the dwell time so that the breakpoint occurs at 40%-50% of the length of the strip chamber, followed by Multiple jet rinses with tap and deionized water.

A third new layer of photoresist, a Laminar 5000 series resist sold by Morton International Electronics Materials, was laminated onto the sheet. The resist layer was covered with an optical tool, and the uncovered parts (all areas except the discrete resistor areas) were exposed with 70 mJ of UV light. Then use a conveyorized spray developer at about 25psi to develop with a 1% sodium carbonate monohydrate solution at 80°C to remove the unexposed resist, adjusting the dwell time so that the break point occurs at 40%-50 of the length of the developing chamber % position, followed by multiple jet rinses with tap and deionized water. The exposed copper areas were then etched in a commercial cupric chloride etchant to remove only the copper leaving the Pt/ _SiO2 exposed and uncorroded to form a resistor connected at each end with copper circuit traces. Use a conveyorized spray resist stripper to remove the photoresist with a 3% NaOH solution at 130 F at about 25 psi, adjusting the dwell time so that the breakpoint occurs at 40%-50% of the length of the strip chamber, followed by Multiple jet rinses with tap and deionized water.

Example 9

Resistors with SiO2 Barriers

This example is an example of how to make a buried resistor using a _SiO2 barrier layer.

Starting with a copper foil of the desired thickness for the final circuit traces, a _SiO2 barrier layer of about 20-50 nm thick is deposited on this copper foil by CCVD deposition. This can be achieved by deposition on a single piece of foil or using a roll process (reel to reel).

After the barrier deposition process, a resistive material (such as Pt metal doped with 2.5% SiO ₂ ) is deposited to a thickness of approximately 100-150 nm using a CCVD process. At this point the quality of the deposited material (its thickness, composition and bulk resistivity) is tested.

The actual resistive material samples consisted of an amorphous silica coating as a barrier layer and a covering resistive layer of platinum-silica composite. The substrate was copper foil measuring 24" x 30" with a coated area of 18" x 24".

The solution of the resistor precursor contained 0.512% by weight of diphenyl-(1,5-cyclooctadiene) platinum (II), 0.028% by weight of tetraethoxysilane, 58.62% by weight of Toluene and 40.69% by weight propane. The solution of the silica precursor contained 0.87% by weight tetraethoxysilane, 8.16% by weight isopropanol and 90.96% by weight propane. Deposition of resistive coatings with silicon dioxide barrier layers also used lower concentrations of Pt( _SiO2 ) precursor solutions, such as 80%, 75%, 65% and 50% of the concentrations mentioned above.

Deposition was performed using a four-nozzle CCVD system as follows: for the first _pass of silicon dioxide, the temperature was 650°C; for the second pass of silicon dioxide, the temperature was 750°C; layer, the temperature is 700°C.

To remove unwanted resistive material from the assembly, a photoimageable resist such as Laminar 5000 is coated over the resistive material. Expose the photoresist material using standard photoprocessing techniques (such as UV exposure through a photomask), and remove unpolymerized photoresist using a suitable solvent (such as 2% sodium carbonate solution at 80°C) to expose the The resistive material is removed during the subsequent lift-off etch. The assembly is then processed through a spray etcher where the assembly is contacted with a glass etchant solution (e.g., 1.7% by weight ammonium bifluoride and 1.05% by weight fluoroboric acid in water) for a sufficient time to chemically etch the _SiO2 barrier and strip off unwanted resistive material. The principle of this method is that the etchant penetrates the micropores in the resistive material to etch the underlying _SiO2 layer. As the glass resist solubilizes the _SiO2 layer, the resistive material loses its cohesion, and because of its thin thickness, the resistive material breaks into small pieces that are carried away as solids by the sprayed etchant material. The contact time with the etchant is limited to a period of time long enough to remove the resistive material but not long enough (about 15-60 seconds) to cause underetching of the desired material covered by the photoresist.

The resistive material is then transferred to a standard epoxy laminate using an industrial lamination process: a sheet of 76289 prepreg is placed on the resistive material side of the etched resistive assembly, followed by an organic release sheet. The assembly was placed on a standard PWB laminator and cured using standard lamination conditions. After lamination, the release sheet is peeled from the laminated assembly to remove the copper to expose the resistors and form the connecting circuit traces. This removal of copper is accomplished using standard photoprocessing processes and etching with copper chloride. The resistor is formed by removing the copper from the upper surface while leaving the copper to connect the two ends of the resistor.

Example 10

Resistors with Nickel Barrier

The following is an example of how to make a buried (buried) resistor using a nickel barrier layer.

Starting with a copper foil of the desired thickness for the final circuit traces, a nickel metal barrier layer of about 2-5 microns thick is deposited on the copper foil by electroplating or CCVD deposition. This can be achieved by deposition on a single piece of foil or using a roll process (reel to reel).

After the barrier deposition process, a resistive material (such as Pt metal doped with 2.5% SiO ₂ ) is deposited to a thickness of approximately 100-150 nm using a CCVD process. At this point the quality of the deposited material (its thickness, composition and bulk resistivity) is tested.

Process actual resistive samples with a nickel barrier. These samples consisted of three 18" x 24" copper sheets that were electroplated with nickel using a commercial nickel plating bath. Three thicknesses of nickel were deposited, approximately 3.5, 7.0 and 10.5 microns. The substrate was a commercial copper foil used in the manufacture of standard PWBs (printed wiring boards).

The resistive material was deposited using a resistive precursor solution containing 0.5 to 12 wt% diphenyl-(1,5-cyclooctadiene) platinum(II), 0.028 wt% ) of tetraethoxysilane, 58.62% (by weight) of toluene and 40.69% (by weight) of propane. Deposition of resistive coatings with nickel barrier layers also used lower concentrations of Pt(SiO ₂ ) precursor solutions, such as 80%, 75%, 65% and 50% of the concentrations mentioned above.

Deposition was performed using a four-nozzle CCVD system using a temperature of 700°C for all three overlays of Pt( _SiO2 ) resistive coating.

To remove unwanted resistive material from the assembly, a photoimageable resist (such as Laminar 5000) is applied to both sides of the resistive material assembly (if a selective etch material that etches nickel but not copper is used) If the resistive material is peeled off, then only one side of the resistive material needs to be coated with a photosensitive resist material). Expose the photoresist material using standard photoprocessing techniques (such as UV exposure through a photomask), and remove unpolymerized photoresist using a suitable solvent (such as 2% sodium carbonate solution at 80°C) to expose the The resistive material is removed during the subsequent lift-off etch. The assembly was then processed through a spray etcher where a commercial cupric chloride etching solution was sprayed on the assembly causing exfoliation etching of the resistive material. The principle behind this method is that the etchant penetrates microscopic pores in the resistive material and corrodes the underlying nickel layer. As the glass resist solubilizes the nickel layer, the resistive material loses its cohesion, and because of its thin thickness, the resistive material breaks into small pieces that are carried away as solids by the sprayed etchant material. The contact time with the etchant is limited to a period of time sufficient to remove the resistive material without being too long (about 15-60 seconds) to etch through the copper foil carrier.

The resistive material is then transferred to a standard epoxy laminate using an industrial lamination process: a sheet of 76289 prepreg is placed on the resistive material side of the etched resistive assembly, followed by an organic release sheet. The assembly was placed on a standard PWB laminator and cured using standard lamination conditions. After lamination, the release sheet is peeled from the laminated assembly to remove the copper to expose the resistors and form the connecting circuit traces. This removal of copper is accomplished using standard photoprocessing processes and etching with copper chloride. The resistor is formed by removing the copper from the upper surface while leaving the copper to connect the two ends of the resistor.

Example 11

Deposition of strontium oxide barrier layer

A strontium oxide coating was deposited on the copper foil using a CCVD process. During deposition, the solution flow, oxygen flow and cooling air flow were kept constant. The solution of strontium oxide precursor contained 0.71% by weight strontium 2-ethylhexanoate, 12.75% by weight toluene and 86.54% by weight propane. At 65 psi, the flow rate of the solution is 3.0 ml/min and the flow rate of oxygen is 3500 ml/min. The cooling air was at room temperature with a flow rate of 25 liters/minute at 80 psi. The cooling air was directed to the back of the substrate with copper tubing, the end of which was placed 2 inches from the back of the substrate. Deposition was carried out at a flame temperature of 700°C as measured at the substrate surface with a type K thermocouple. The cooling air flow may be in the range of 15-44 liters/minute. The deposition temperature was varied from 500-800°C.

Example 12

Deposition of zinc oxide barrier layer

A zinc oxide coating is deposited on the copper foil using a CCVD process. During deposition, the solution flow, oxygen flow and cooling air flow were kept constant. The solution of the zinc oxide precursor contained 2.35% by weight zinc 2-ethylhexanoate, 7.79% by weight toluene and 89.86% by weight propane. At 65 psi, the flow rate of the solution is 3.0 ml/min and the flow rate of oxygen is 4000 ml/min. The cooling air was at room temperature with a flow rate of 25 liters/minute at 80 psi. The cooling air was directed to the back of the substrate with copper tubing, the end of which was placed 2 inches from the back of the substrate. Deposition was carried out at a flame temperature of 700°C as measured at the substrate with a type K thermocouple. The flow rate of cooling air may be in the range of 9-25 liters/minute. The deposition temperature was varied from 625-800°C.

Example 13

Deposition of tungsten oxide barrier layer

A tungsten oxide coating is deposited on the copper foil using a CCVD process. During deposition, the solution flow, oxygen flow and cooling air flow were kept constant. The solution of the tungsten oxide precursor contained 2.06% by weight tungsten hexacarbonyl, 26.52% by weight toluene and 73.28% by weight propane. At 65 psi, the flow rate of the solution is 3.0 ml/min and the flow rate of oxygen is 3500 ml/min. No cooling gas was used at the deposition temperature of 350°C. The temperature is measured at the surface of the substrate with a type K thermocouple. A flow of cooling gas in the range of 7-10 liters/minute can be introduced to the back of the substrate during deposition. The deposition temperature can vary from 350-800°C.

Claims (17)

1. three-decker that is used to form discrete resistors, it comprises:

The layer of metal conductive layer,

One deck intermediate layer, by can be formed by the etched material of chemical etchant and

The resistance elements that one deck is made of resistance material, it has enough porositys and can permeate the described intermediate layer of chemical etching by described resistance material so that be used for the described chemical etchant in described intermediate layer, thereby described resistance material can be peeled off by the described conductive layer in the place of chemical etching from those described intermediate layers.

2. three-decker as claimed in claim 1, wherein said intermediate layer has barrier function, enters in the described resistance elements from the diffuse of described conductive layer in order to preventing.

3. three-decker as claimed in claim 1, wherein said intermediate layer is a metal.

4. three-decker as claimed in claim 1, wherein said intermediate layer is a nickel.

5. three-decker as claimed in claim 1, wherein said intermediate layer is a dielectric materials layer, its average thickness is the 15-50 nanometer.

6. three-decker as claimed in claim 1, wherein said intermediate layer is selected from: silicon dioxide, strontium oxide strontia, tungsten oxide or zinc oxide.

7. the one deck that electrically contacts with layer of conductive material is through the formation method of the resistance material of composition, and it comprises:

Provide as each described three-decker in the claim 1 to 6,

On described resistance elements, form the photoresist layer of one deck through composition,

Described resistance elements is contacted with the described chemical etchant that is used for described intermediate layer, so that described etchant permeates by the resistance elements of described porous the described intermediate layer of chemical etching, and

Peel off in those etched places, described intermediate layer and to remove described resistance material layer segment.

8. method as claimed in claim 7, wherein said etchant are selected from bifluoride hydrogen ammonium, fluoboric acid, hydrochloric acid and their mixture.

9. structure comprises:

A) layer of conductive material and

B) the bonding resistance material thereon of one deck, described resistance material have enough porositys so that liquid etchant sees through its diffusion and destroys described a) layer and b) bonding between the layer.

10. structure as claimed in claim 9, wherein said a) layer is a metal forming.

11. structure as claimed in claim 9, wherein said a) layer is a Copper Foil.

12. structure as claimed in claim 9, wherein said resistance material are the metals that is doped with dielectric material.

Platinum or nickel 13. structure as claimed in claim 12, wherein said resistance material have been mixed.

14. structure as claimed in claim 12, wherein said resistance material is a conductive oxide.

15. structure as claimed in claim 12, the thickness of wherein said resistance material are 40-50,000 dust.

16. a method that forms on the insulation ground through the composition resistance material comprises:

The resistance elements that provides one deck of being positioned on the described insulating material to constitute by resistance material, described resistance material has enough porousness, so that can seeing through described resistance elements diffusion, etchant weaken bonding between described resistance material and the described insulation ground

Cover with photoresist described resistance elements selected part and

Allow unmasked portion on the described resistance elements contact with etchant to destroy bonding between described resistance material and the described insulating material.

Platinum or nickel 17. method as claimed in claim 16, wherein said resistance material have been mixed.

Images